INFORMATION TO USERS
The most advanced technology has been used to photo graph and reproduce this manuscript from the microfilm master. UMI films the original text directly from the copy submitted. Thus, some dissertation copies are in typewriter face, while others may be from a computer printer.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyrighted material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are re produced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each oversize page is available as one exposure on a standard 35 mm slide or as a 17" x 23" black and white photographic print for an additional charge.
Photographs included in the original manuscript have been reproduced xerographically in this copy. 35 mm slides or 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
Accessing theUMI World’s Information since 1938
300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA
Order Number 8824495
Gender relations and the suburbanization of clerical work
England, Kim V. L., Ph.D.
The Ohio State University, 1988
Copyright ©1988 by England, Kim V. L. All rights reserved.
UMI 300 N. ZeebRd. Ann Arbor, MI 48106
PLEASE NOTE:
In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identif ied here with a check mark V .
1. Glossy photographs or ______pages
2. Colored illustrations, paper or ______print
3. Photographs with dark background_____
4. Illustrations are poor copy______
5. Pages with black marks, not original ______copy
6. Print shows through as there is text on both sides ______of page
7. Indistinct, broken or small print on several pages _
8. Print exceeds margin requirements______
9. Tightly bound copy with print lost in______sp in e
10. Computer printout pages with indistinct______print
11. Page(s)______lacking when material received, and not available from school or author.
12. Page(s) seem to be missing in numbering only as text follows.
13. Two pages num bered . Text follows.
14. Curling and wrinkled pages
15. Dissertation contains pages with print at a slant, filmed as received
16. Other
UMI
GENDER RELATIONS AND THE SUBURBANIZATION OF CLERICAL WORK
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the Graduate
School of The Ohio State University
By
KIM V.L. ENGLAND, B.A.(Hons), M.A.
*****
The Ohio State University
1988
Dissertation Committee: Approved by
Kevin R. Cox
Lawrence A. Brown
W. Randy Smith Kevin R. Cox, /(dviser Department of Geography Copyright by KIM V.L. ENGLAND 1988 DEDICATION
For
Florence England and George Forster
and
Muriel England and Stan England ACKNOWLEDGEMENTS
First and foremost I wish to thank my parents, whose love and support have sustained me and brought me to this point. For the last six years they have encouraged me from afar. As a result of their efforts I have been the first person in my extended family to go to university, although
I very much doubt that they expected me to be at university for a few months shy of a decade (however their bank account can verify this). I am grateful to my parents for giving me things they never had and encouraging me to take opportunities they wish they had had, and for always let ting me know that they are there for me.
I would like to express my sincere gratitude to Kevin
Cox, my advisor, who has taught me many things: how to read, write, and think like a professional social scien tist. I admire Kevin for his tireless enthusiasm for research and his remarkable dedication to his students. I consider myself lucky to have been a recipient of this ded ication in the last six years.
Special thanks go to Larry Brown for seeing potential in me at a time when few others did. He has helped me to build my self-confidence, to believe in myself and take myself seriously. It was Larry who put me on the road to
Feminist Geography, assuring me that my personal qualities were valid for research purposes. For these gifts I am grateful.
I would also like to thank Randy Smith, who has been involved with my progress and research for most of my career at Ohio State. Randy's insights into my research and sustaining encouragement will always be remembered with great fondness.
The financial support of the Geography Department is much appreciated. I am grateful to the Urban Affairs and
Urban Assistance Program, which funded the grant on which this dissertation is based. And I want to thank the Center for Women's Studies for not only allowing me the privilege of teaching introductory classes in Women's Studies, but even paying me for it too.
I am deeply grateful to the twenty-eight women who took time out of their busy lives in order to share their experiences with me. Their conversations with me, excerpts of which appear in this dissertation, were both enriching and inspiring. Equally, I want to thank the ten personnel managers whom I interviewed for their important insights into the "real world" of my "concrete" research.
- iv - Dissertation writing and research are often difficult tasks, but they would have been a good deal more difficult for me without the following people in Geography: Frank
Borgers, Paul Burke, Ayse Can, Alan Edmonds, Elena Kambit- si, Vicky Lawson, Kavita Pandit, Monica Short, Darlene Wil cox, Enid Wray and Jenny Zorn. And from Women's Studies:
Mary Margaret Fonow, Susan Hartmann, and Clare Robertson.
My sanity has remained almost intact because of my rela tionships with some wonderful women: Christine Hodgkinson, my first true woman friend (after my mother); Kim Kaplan, kindred spirit and source of so much strength during some of my "darkest hours"; Lisa Mutschler, witty and intelli gent, brightening many dull moments; Sue Roberts, fellow geographer and friend since our undergraduate years togeth er; and a special man: Michael Dayton, a constant source of relief and happiness in the last three years. Michael has always helped me realize that there is indeed more to life than Geography, Women's Studies and my dissertation.
Last and by no means least I wish to thank Frank Borgers and Kmel for their love, support and understanding over the last couple of years and particularly in the last few months. While one of them has remained a silent critic of my work, the other has not and his input into my research and life have been truely irreplaceable.
- v - VITA
February 16, 1960 ...... Born, Newcastle-upon-Tyne, England.
1981 ...... B.A. (Honours), Combined Arts (Geography, Sociology, History and Art History) University of Leicester, Leicester, England.
1982 ...... P.G.C.E. (Post-Graduate Certificate of Education) Geography, School of Education, University of Leicester, England.
1984 ...... M.A., Geography, The Ohio State University, Columbus, Ohio.
1982-1986 ...... Graduate Teaching Associate, Department of Geography, The Ohio State University, Columbus, Ohio.
1986-1987 ...... Graduate Research Associate, Urban Affairs and Urban and Urban Assistance Program, The Ohio State University, Columbus, Ohio.
1987-1988 ...... Graduate Teaching Associate, Center for Women's Studies, The Ohio State University, Columbus, Ohio.
PUBLICATIONS
"The Practical Adequacy of 'METHOD IN SOCIAL SCIENCE'" ANTIPODE, 1986: 240-254 (with A. Jonas and A. Mair).
"Education and Residence in Historical Perspective",
- vi - 1988, forthcoming (with K. R. Cox).
"Location, Social Categories and Labor Force Experiences in Developing Economies: The Venezuelan Case" THE INTERNATIONAL REVIEW OF REGIONAL SCIENCE, 1988, (with L. A. Brown and A. R. Goetz).
"The Employment of Suburban Women in Ohio" GEO-SPECTRUM, 6, 1986: 6-8.
"Gender, Location, and Human Resource Experiences in Developing Economies: The Venezuelan Case" Discussion Paper Number 32, in STUDIES ON THE INTERRELATIONSHIPS BETWEEN DEVELOPMENT AND MIGRATION IN THIRD WORLD SETTINGS, DISCUSSION PAPER SERIES, Department of Geography, The Ohio State University, Columbus, Ohio, 1986 (with L. A. Brown and A. R. Goetz).
"Education and Residential Utility, 1940-1980" GEO-SPECTRUM, 4, 1984: 7-8.
FIELDS OF STUDY
Major Field: Social Theory and Geography
Studies in Social Theory and Geography: Professor Cox.
Studies in Urban Geography: Professors Cox, L. A. Brown, and Smith.
Studies in Development: Professor L. A. Brown. TABLE OF CONTENTS
PAGE
D e d i c a t i o n ...... ii
Acknowledgements ...... iii
VITA ...... vi
LIST OF FIGURES ...... ix
LIST OF T A B L E S ...... x
CHAPTER
I. INTRODUCTION ...... 1
1.1 Women's Paid Employment Since the Second World War ...... 5 1.2 Feminist Analysis and Feminist G e o g r a p h y ...... 12 1.3 The Social Theory and Geography M o v e m e n t ...... 14 1.4 Organization of this Dissertation . . . 19
II. THE LITERATURE: QUESTIONS OF SOCIAL STRUCTURE . 22
2.1 Introduction...... 22 2.2 Patriarchy and C a p i t a l i s m ...... 23 2.2.1 The Debate ...... 24 2.2.2 Gender Division of Labor within the Labor Force ...... 30 2.2.3 Gender Division of Labor within the Household ...... 38
- viii - 2.3. Patriarchy and Capitalism in Historical Perspective ...... 45 2.3.1 The Changing Character of M o t h e r h o o d ...... 46 2.3.2 Women in "Non-Traditional" H o u s e h o l d s ...... 49 2.3.3 The Redefining of Children .... 53 2.3.4 Household Technology ...... 55 2.3.5 Legislation and Popular Attitudes Affecting Women's Paid W o r k ...... 59 2.4 The French Regulation School ...... 64 2.4.1 Capital and Clerical w o r k ...... 69 2.5 The Collapse of F o r d i s m ...... 80 2.5.1 Declining Real Wages and the "Middle-Class Dream" ...... 81 2.5.2 "Flexibility": Temporary and Part-Time W o r k ...... 87 2.6 S u m m a r y ...... 93
III. THE LITERATURE: QUESTIONS OF S P A C E ...... 95
3.1 Introduction...... 95 3.2 Spatial Entrapment: A Conceptualization ...... 96 3.3 Implications of Time-Space Budgeting . . 101 3.3.1 Mobility and Gender ...... 101 3.3.2 The Human Built Environment and G e n d e r ...... 102 3.3.3 Residential Location and Gender . . 108 3.4 Spatial Entrapment and Female Clerical Employment...... 110 3.5. Evaluating Spatial Entrapment: Mobility Studies ...... 118 3.6 Summary ...... 122
IV. METHODOLOGY ...... 124
4.1 Introduction...... 124 4.2 Contrasting Approaches to Theory .... 124 4.3 Contrasting approaches to Methodology . 129 4.4 Methodology Employed In This Dissertation ...... 135 4.4.1 Extensive Research ...... 135 4.4.2 Intensive Research ...... 136 4.5 Summary ...... 137
- ix - V. WOMEN'S MULTIPLE ROLES AND SPACE: A QUANTITATIVE ANALYSIS ...... 139
5.1 Introduction ...... 139 5.2 Spatial Distribution of Female Clerical Workers ...... 141 5.3 Multiple Regression Analyses ...... 158 5.4 Commuting In C o l u m b u s ...... 165 5.4.1 Polk's Suburban Directory Data S e t ...... 166 5.4.2 Personnel Files D a t a ...... 169 5.4.3 Survey Questionnaire D a t a ...... 175 5.5 S u m m a r y ...... 183
VI. WOMEN'S MULTIPLE ROLES AND SPACE: A QUALITATIVE ANALYSIS ...... 187
6.1 Introduction...... 187 6.2 Interview D a t a ...... 189 6.3 The Spatial Entrapment Thesis ...... 192 6.4 An Alternative Explanation of Women’s Commuting Patterns ...... 196 6.4.1 Divorced W o m e n ...... 197 6.4.2 Domestic arrangements ...... 199 6.4.3 Career aspirations ...... 203 6.4.4 Work Schedules ...... 206 6.5 Societal Changes ...... 208 6.5.1 Child-care Facilities ...... 208 6.5.2 Employer Flexibility ...... 210 6.6 D i s c u s s i o n ...... 211 6.7 Office Location and Spatial E n t r a p m e n t ...... 217 6.8 Summary ...... 220
VII. SUMMARY AND C O N C L U S I O N S ...... 223
7.1 Summary of the Major Findings ..... 223 7.2 Contributions of this Dissertation . . . 228
APPENDICES
A. INTERVIEW ITEMS FOR INTERVIEWS WITH WOMEN . . . 231
- x - B. INTERVIEW ITEMS FOR PERSONNEL MANGERS ...... 244
C. ADDITIONAL STATISTICS FROM THE REGRESSION ANALYSES ...... 247
D. SURVEY QUESTIONNAIRE ...... 249
Bibliography ...... 253 LIST OF FIGURES
FIGURE PAGE
1. Jane's Daily Prism...... 98
2. Franklin County, Shares of Female Clerical Workers, 1950 (Map 1)...... 143
3. Franklin County, Shares of Female Clerical Workers, 1950 (Map 2)...... 144
4. Franklin County, Shares of Female Clerical Workers, 1960 (Map 3)...... 145
5. Franklin County, Shares of Female Clerical Workers, 1960 (Map 4)...... 146
6. Franklin County, Shares of Female Clerical Workers, 1970 (Map 5)...... 148
7. Franklin County, Shares of Female Clerical Workers, 1980 (Map 6)...... 149
8. Franklin County, Location Quotients, 1950 (Map 7)...... 151
9. Franklin County, Location Quotients, 1950 (Map 8) ...... 152
10. Franklin County, Location Quotients, 1960 (Map 9 ) ...... 153
11. Franklin County, Location Quotients, 1960 (Map 10)...... 155
12. Franklin County, Location Quotients, 1970 (Map 11)...... 156
13. Franklin County, Location Quotients, 1980 (Map 12)...... 157
- xii - LIST OF TABLES
TABLE PAGE
1. Labor Force Participation of Women by Age.... 6
2. Labor Force Participation of Ever-Married Women with Children ...... 7
3. Ohio: Percent in the Labor Force, 1940-1980 . . . 9
4. Percent in the Labor Force, 1970-1980 10
5. Percent of all Women in Paid Labor in each Occupation Class ...... 31
6. Percent of Workers who are Women in each Occupational Class ...... 33
7. Median Earnings of Women and Men by Occupation, 1983...... 35
8. Changing Composition of Households: 1940 to 1980 ...... 51
9. Changes in Personnel Practices Resulting from EEO L a w s ...... 61
10. Growth of the Clerical Labor Force, 1870-1980...... 69
11. The Feminization of Clerical Work, 1870-1980 . . 72
12. Male and Family Median Income Growth, 1949 to 1984 ...... 83
13. Spouse in the Paid Labor Force, 1950-1980 . . . 85
14. Median Family Income: Presence of Wife in Labor F o r c e ...... 86
15. Summary of Journey-to-Work studies: mean miles (and minutes) ...... 119
- xiii - 16. A Summary of Intensive and Extensive Research . 134
17. Independent Variables Employed in Regression A n a l y s e s ...... 160
18. Regression Coefficients (Betas) ...... 163
19. Commuting Distances of Female Clerical Workers and their Husbands...... 167
20. Commuting Estimates for Firm l's Female Clerical Workers ...... 170
21. Commuting Estimates for Firm 2's Clerical W o r k e r s ...... 171
22. Commuting Estimates for Firm 3's Female Clerical Workers ...... 172
23. Demographic Profile of Questionnaire Respondents...... 177
24. Journeys-to-Work Distances and Times by Marital Status ...... 178
25. Journey-to-Work Distances and Times by Parental S t a t u s ...... 178
26. Journey-to-Work Distances and Times ...... 180
27. Profile of the Interviewees ...... 189
28. Means and Standard Deviations of Variables E m p l o y e d ...... 247
29. Zero Order Correlations ...... 248
- xiv - CHAPTER I
INTRODUCTION
I began work in 1943 as a clerk for (a downtown insurance firm). My dad was real pleased I was making something of myself, being from a farming background. He thought that office-work was real grand that I didn't go into domestic service like most farm girls. I rented an apartment on South High Street and took the street car. Oh, everyone took the street car to work, the only people who drove came in from out of town, and even then they car-pooled. Not like now, when I go into the different offices 'round town (she works for a temporary agency) everyone drives and on their own too (Betty W ) .
I worked on a schedule where I didn't need to be at work until 8:45. The school bus came at 8:15, I could drop them off at the bus stop or at school. They got home from school at 4:15, and it was forty-five minutes before I got home. But there was a neighbor who didn't work, and I made arrangements with them where they'd fetch the boys if they got sick. You need to make arrange ments. We all worked together on this, we traded off. Sometimes the neighbors wanted to got out, so I'd watch their kids (Sally).
These quotes by women clerical workers provide insights into the lived reality of women who are in paid employment.
In her reminiscing about working as a clerical worker for most of the last forty-five years, the first woman illus trates the changing geography of clerical work. The reason
"everyone took the street car to work" was that it took
- 1 - them downtown where the majority of clerical jobs were located at that time. Today, as a temporary worker she finds herself working all over the city, getting to those places by driving alone in her car. The second woman high lights the sheer complexity of women's coping strategies when they combine their "domestic" roles (including those of mother, domestic manager, and partner) with the role of waged worker. As a single parent, this woman had a height ened awareness of the necessity for women with multiple roles to budget their time carefully.
These are two of the central themes of this disserta tion, which explores the links between changing spatial patterns of clerical work and the rapid increase in women's paid employment since 1945. In exploring these links attention will initally focus on explanations, related to social and spatial structures, which account for the post-
Second World War increase in women as waged laborers in general, and in women clerical workers in particular. Of particular interest in this dissertation are the existing explanations for the changing spatial patterns of clerical work at the metropolitan scale. Until about the 1960's most clerical jobs were located downtown, but, as a number of scholars have argued, since the 1960's firms have been relocating the routine aspects of their clerical functions from city centers to the suburbs. It is contended that besides cheaper land and rent costs, the attraction of sub urban locations is their female labor supply (Alexander,
1979; Baran, 1985, 1987; Nelson, 1986a, 1986b; Sassen-Koob,
1984; White, 1977; 9-to-5, National Association of Working
Women, 1985). Three suppositions appear to underlay the explanations: first, due to their commitment to their domestic roles suburban women can only support short journeys-to-work and so are a "captive labor force, spa tially entrapped in their area of residence; second, due to owning newly constructed housing and the existence of young children, many married suburban mothers need to be in paid employment; and third, due to the "preferred clerical labor supply" characteristics of suburban women (low-waged, well- educated, non-militant, married women with young children), firms employing large numbers of female clerical workers
(and which wish to accrue savings from employing an educat ed but cheap female clerical labor force) may choose to relocate to suburban residential areas of new single-family housing and of concomitantly spatially entrapped women located within easy access of highway facilities.
This dissertation investigates the applicability of these explanations for the case of the Columbus metropoli tan area by employing both quantitative and qualitative methods. Specifically the quantitative analyses first explore whether, in fact, there has been a suburbanization of female clerical labor in the Columbus metropolitan area; ; the relationships between the growth of clerical work in suburban locations and measures of the need of married sub urban mothers for paid employment (such as home-ownership); and if there are relations between the growth of female clerical workers, measures of need and suburban locations, do they have anything to do with the spatial entrapment of women and, hence, the recent arrival of offices? The sec ond part of the quantitative analysis involves an examina tion of the journeys-to-work of women clerical workers who live and/or work in a number of Columbus' suburbs. A variety of data is employed in an effort to ascertain whether female clerical workers are in fact spatially entrapped.
Qualitative methods are employed in this dissertation in an interpretation of the relationships revealed by the quantitative analysis This is achieved through the use of the unstructured, interactive interviews which were con ducted with twenty-eight female clerical workers and with ten personnel managers of firms which employ a large number of clerical workers. Essentially these interviews high light the spuriousness of linking women's short commutes with heavy domestic reponsibilities. Thus an alternative explanation of the women's journey-to-work patterns to that
offered by the spatial entrapment thesis is presented. In this alternative explanation attention focusses on how the women’s commutes reflect their changing powers and needs, the spatial patterns of job opportunities, and the extent to which they are enmeshed in a pre-existing set of local ized relations. The results of interviews with the person nel managers of ten firms are also presented, paying spe cial attention to the spatial aspects of their labor markets.
1.1 WOMEN'S PAID EMPLOYMENT SINCE THE SECOND WORLD WAR
Since the Second World War the participation of women in the paid labor force has increased significantly. Few women in the United States worked outside the home at the turn of the century (between 1900 and 1940 women made up about 20% of the paid labor force and only about 20% of all women of employable age were in paid employment). In 1950 women made up 29.6% of the paid labor force, and 33.9% of all women of employable age were in paid employment. By
1980, however, the equivalent numbers were 43.5% and 52.9%.
Indeed in the past thirty years, six out of ten additions to the paid labor force have been women.1
These aggregate level figures hide two significant post-
Second World War trends in the employment patterns of women
in the paid labor force. First, much of the initial
1 Between 1950 and 1980, male participation rates have declined by approximately ten percentage points. 6 post-1945 increase in female paid employment consisted of older women, presumably re-entering the labor force once their children had grown (Table 1). In 1960 46.7% of women aged 45 to 54 were in the paid workforce, an increase of
13.8 percentage points since 1950, in 1970 their rate was
52.5%. Of women aged 55 to 64, their rate increased 11.5 percentage points between 1950 and 1960 (from 23.5% to
35.0%); and in 1970, 42.1% of them were in paid employment.
The significance of this influx of older women into the labor force during the 1950's is particularly apparent when the increases are compared to the 5.8 percentage points increase for all women (from 29.9% in 1950 to 35.7% in
1960).
Table 1
Labor Force Participation of Women by Age
1950 1960 1970 1980
All 29.9 35.7 41.4 49. 9 16-19 31.1 32.9 34.9 45.8 20-24 42.9 44.8 56.1 67.8 25-34 31.8 35.3 44.9 64.8 35-44 35.0 42.7 50.3 64.7 45-54 53.9 46.7 52.5 58.8 55-64 23.5 35.0 42.1 41.6 65+ 7.8 10.2 10.0 8.2
Source ! U.S. Bureau of the Census
Second, the more recent trend has been for young women
to remain in the labor force during their "prime child bearing years" (Table 2). Indeed women with children aged 7
Table 2
Labor Force Participation of Ever-Married Women with Children
Under Age Age Under YEAR AGE 3 3 TO 5 6 TO 17 Age 18
1950 (13.6% under age 6) 32.8 21.6 1960 16.5 27.4 42.5 30.4 1970 25.7 39.2 51.5 42.0 1975 34.4 44.6 54.8 47.4 1980 41.7 54.5 64.3 56. 7
Sources U.S. Bureau of the Census. less than six are the fastest growing sub-group of women entering the paid work-force today (Bianchi and Spain,
1986). For instance in 1950 while only 13.6% of mothers with children aged less than six were in paid employment,
32.8% of mothers with children aged 6 to 17 worked. By
1980, about half of all mothers with young children worked, compared with 64.3% of mothers of older children.2
At the same same time patterns of women's paid employ ment have varied spatially. Indeed it is possible to iden tify fairly regular geographical trends in the geography of women's paid employment since the Second World War. While the number of women in paid employment has increased throughout metropolitan areas, the rate has been particu larly significant in suburban areas, especially since 1960
(Hanson and Hanson, 1980; Saegart, 1981).
2 This trend has continued through the 1980's. In 1985 54.0% of mothers with children aged less than six were in waged work (Ohio Senate Task Force on Women Single Heads of Households, 1985). These trends can be illustrated using 1940-80 Census data for five urban areas in Ohio, and using County remain ders as proxies for the suburbs. Table 3 shows the percent of women who participated in the labor force between 1940 and 1980. These results reflect national trends, in that female participation rates have increased throughout metro politan areas but particularly in suburban areas, and com monly at an increasing rate since 1960. For instance in
Dayton women constituted 29.8% of the labor force in 1940, while in Montgomery County the number was only 19%. These rates increased over the period, and by 1980 the participa tion rate of 52% in the County remainder was greater than that in the city, where the rate was lower at 48%. Indeed this trend reversal over time typifies every case in this study. Table 3
Ohio: Percent in the Labor Force, 1940-1980
1940 1950 1960 1970 1980 Cleveland male 81.2 81.9 81.1 75.1 67.5 female 30.6 35.0 36.6 43.0 45.0
Cuyahoga male 81.1 84.6 80.6 79.3 79.5 female 27.2 29.2 36.6 42.6 51.4
Cincinnati male 79.6 78.2 79.3 72.3 70.0 female 29.8 33.1 34.5 43.4 48.8
Hamilton male 80.5 82.2 79.4 77.6 81.2 female 22.6 38.3 35.0 41.1 49.9
Toledo male 79.2 81.2 79.6 78.1 74.0 female 28.9 32. 7 29.8 42.4 48. 7
Lucas male 80.1 83.6 79.9 78.9 81.2 female 18.3 24.1 35.0 41.5 49.9
Dayton male 80.4 82.7 81.5 73.7 67.5 female 29.8 35. 7 37.3 48.9 48.0
Montgomery male 72.4 80.0 81.2 80.4 80.0 female 19.0 25.7 37.0 43.4 52.0
Columbus male 75.4 74.5 76.1 75.2 64.6 female 30.2 36.2 40.6 47.3 55.2
Franklin male 76.4 78.5 83.5 84.3 66.5 female 20.3 26.0 32.4 44.5 55.3
Source: U.S. Bureau of the Census 10
Table 4
Percent in the Labor Force, 1970-1980
MALE FEMALE 1970 1980 1970 1980
Columbus 75.2 73.9 47.3 55.4
Gahanna 89.4 86.2 45.2 56.1
Reynoldsburg 89.6 87.5 46.0 64.5
Upper Arlington 81.4 80.5 35.9 46.9
Westerville 78.1 84.4 42.8 52.1
Worthington 85.4 84.5 38.8 54.3
Source: U.S. Bureau of the Census
Similar data are available for Columbus and a number of its suburbs, but only for 1970 and 1980. However these data (Table 4) reveal a similar patterning of changes in participation rates. In some suburbs (Worthington and Rey noldsburg) the increase in women's employment has been quite dramatic, while the city of Columbus has the lowest rate of change.
Despite the large numbers of women who have been enter ing the paid labor force for the first time, however, women are concentrated into a smaller number of occupations com pared to men. That half of all women in paid employment are located in just twenty-one occupations, whereas half of men may be found in a total of sixty-five occupations suc cinctly illustrates this aspect of the gender division of 11 labor within the paid labor force (Coser, 1982). Indeed, of all women working for pay in the United States one third work in just one occupation: clerical work. This has prompted Glenn and Feldberg (1984) to describe clerical work as the "prototypical female employment experience".
Clerical employment has also varied spatially at the metropolitan level. From the late nineteenth century to about the Second World War most offices were downtown employing mainly young, white, native-born, relatively well-educated women, who worked until they got married
(Davies, 1975, 1982; Kennedy, 1979; Kocka, 1980; Sanderson,
1986). After the Second World War this labor supply became less available and was supplemented by older women past their child-bearing years. But by the 1960's and 1970's both these supplies of labor became increasingly scarce downtown. Today's central city female labor force is com prised of low-income women (many of whom are minorities) and high-income, career-oriented women (Nelson, 1986a).
It is suggested that compared to central city women, suburban women are more likely to be white, well-educated, married and to have young children. The latter two charac teristics apparently make suburban women less likely to be career-oriented, which is desirable for firms wishing to
hire a labor force for dead-end, low paid clerical work. A
central premise of all these studies is that firms seek to 12 minimize labor costs by locating near "pools" of cheap, reliable unorganized female labor spatially entrapped in their area of residence.
This dissertation also examines the changing geography of female clerical work at the metropolitan level. How ever, rather than focussing on the cost-minimizing princi ple of the relocating offices, this dissertation draws on the concepts of patriarchy and gender relations (the rela tions of power between women and men) to go beyond the lev el of appearances in order to understand the changes in employment patterns of female clerical work and the link
(if any) with the recent suburbanization of routine office functions. By paying particular attention to conceptualiz ing how and why these changes have occurred, this disserta tion draws on the insights of two broad bodies of litera ture which have recently been gaining currency in geography: feminist analysis and social theory.
1.2 FEMINIST ANALYSIS AND FEMINIST GEOGRAPHY
A central concern for feminists has been to expose the contrast between the lives of women (traditionally focussed on the "private" sphere of home, family and domesticity); and the lives of men (centered on the "public" sphere of waged work and political activity). Generally the recogni tion of these two socially constructed spheres has been 13 interpreted as meaning that women and men are equal but different. Feminists argue that the differences in the social position of women and men operate to the advantage of men so that in reality women and men have unequal power, opportunities and social prestige (Women and Geography
Study Group of the IBG, 1984).
Geography has, relative to other social sciences, been rather slow in articulating a feminist perspective, but as
Jan Monk (quoted in Gruntfest, 1988) notes:
Since its appearance in the 1970's as an academic expression of the women's movement, feminist geography has become more complex in the ques tions it asks and the approaches it employs, reflecting an increasingly sophisticated scholar ship (1988:1).
Feminist geography, then, is a geography explicitly con cerned with this socially constructed division of gender, as well as a geography committed to the alleviation of gen der inequality in the short term and towards it's ultimate eradication through social change (Women and Geography
Study Group of the IBG, 1984). Thus feminist geographers, in keeping with other feminist social scientists, regard the analysis of gender relations as fundamental to an understanding of most aspects of human activity, as gender relations infiltrate every area of social life. By ignor ing gender divisions, a significant element of the analysis of the causal relationship between human actions and social and spatial structures is missing. As Fincher (1987) points out: 14
the social construction of gender roles cannot be overlooked as an important causal factor in many circumstances, and an important facet of our lived experience and attribution of meaning in many circumstances. This situation is a matter of complete analysis, not of politics (1987:11).
1.3 THE SOCIAL THEORY AND GEOGRAPHY MOVEMENT
This section focuses on the ontological position associ ated with the Social Theory in Geography Movement, and com ments on the methodological implications of this ontologi cal position. A number of geographers have been employing social theories in an attempt to overcome the dualism of society and space. It is suggested that in order to under stand space we have to understand the social relations of people. Thus space is viewed as a social construct; but at the same time social relations are necessarily spatial. In short spatial structures are conceptualized as being both the medium and outcome of social action (hence the duality of society and space.
However this is not to imply a socially deterministic view of spatial structures. Indeed the proponents of the
Social Theory in Geography Movement are equally as con cerned with abolishing the individual/society dualism as
they are that of the dualism of society and space. Hence the conceptualization of the duality of individual and
society: individuals are active only in and through social
structures, but those social structures would not exist 15 without individual activity as they are the reproduced and transformed outcome of individual activity. By putting all these ideas together (individual, society and space) we arrive at the duality of individuals and socio-spatial structures.
In line with its critical approach to the ontological assumptions of mainstream social science the Social Theory in Geography Movement has been concerned about employing methodologies which are sensitive to the nature of the objects being studied. In this regard it has vigorously argued the case that the objects of social science are quite different from those of natural science and as such require a methodology reflecting this fact. In other words a research method appropriate to the natural sciences is not necessarily appropriate for the social sciences. Meth ods employed by natural sciences have tended to be borrowed by social scientists interested in producing empirical reg ularities with the aim of predicting human activity. How ever the production of empirical regularities, necessary for successful prediction, requires two preconditions.
First, the internal structure of the object of study must always be consistent (Bhaskar refers to this as "intrinsic conditions for closure"); second the external conditions must never vary ("extrinsic conditions for closure"). If a
system satisfies both of these conditions it is a closed 16 system, if not, it is an open system. Successful pre diction is only possible in closed system and while some natural scientists deal with closed systems (the consistent outcomes of which can be successfully predicted), social scientists do not.
It is in this context that realism has proven so attrac tive to the proponents of the Social Theory in Geography
Movement. Realism is based on the use of rational abstrac tions to identify the necessary causal powers and causal liabilities of specific social objects manifested under contingent (specific) conditions. To explicate this defi nition an understanding of the realist approach to causa tion is required.
Causation is understood as the necessary ways-of-acting or mechanisms of an object which are derived from its structure. These mechanisms or necessary ways of acting, are in turn conditional upon: causal powers -- objects have the power or need to make things happen; and causal liabilities (or susceptibilities) — objects can suffer from or are susceptible to the activities of other objects
(Sayer, 1984). These powers and liabilities exist even when they are not being exercised or suffered (and before they are discovered); and when they are exercised or suf fered their resultant effect is dependent upon the contin gent conditions (the specific circumstances of time and place) under which they were exercised/suffered. 17
For social objects (such as people) the structure is social in character and so the ensuing causal properties of people largely derive from their social relations with oth er people. Thus, people in a wage labor society have the power to undertake waged labor but whether and with what effects this power is exercised will depend on contingent conditions such as the availability of paid employment.
Thus, while abstract theory can identify those needs and susceptibilities which are necessary properties of their objects, it can only make suggestions about what actually happens in the concrete.
Doreen Massey's work on economic and spatial restructur ing in Britain (1983, 1984, 1987) provides an example of using social theory and realism in here concrete research.
For her, space and social activity are inseparable. Spa tial structures are not only the outcome of social activi ty, but they are also the necessary precondition for that social activity. There is also a notion in her work that an observed spatial outcome could be the result of any num ber of social activities; at the same time the same social activity can produce many result in different spatial out comes depending on the nature of the pre-existing spatial structure.
Ruth Fincher (1987) constructs a research agenda for the
intertwining of social theory and urban geography. She 18 suggests that a complete explanation of urban phenomena would:
(try) to explain human actions with reference to a range of historically specific societal charac teristics which the relevant actors have drawn on actively and which they experience. For example, one could explain the urban renewal decisions of policy makers in terms of locally experienced fiscal crisis, where the relevant aspects of local fiscal crisis are theoretically informed or developed by the researcher, not just derived from statements of the decision makers. Note that voluntarism is avoided, but that the deci sion makers' experience of their context is stressed ... (and try) ... to explain the causal context of a set of activities, emphasizing the theorization of the context rather than the mean ing attributed by actors to that context or their experience of it. Here the appearance of certain societal characteristics as causal in a given circumstance will be stressed and phenomena to be regarded as human actions as a specific selective response to a spatially varying decline in the rate of profit. The explanation would need to emphasize why the pattern of plant closings was more closely related to variations in the rate of profit than to other contextual factors such as the tendency for the rate of profit to be equal ized across regions by state tariff and pricing mechanisms, or why plant closings were the action taken, not the purchase of new technology, in efforts to improve profit rates. The emphasis is on explaining why certain elements of the context come to be causal in particular situations (1987:9) .
In her "manifesto" Fincher states that feminist analysis is one arena in which the union of social theory and geog raphy would flourish. She provides a number of examples of how an emphasis on the social construction of gender in urban geography would generate several sorts of new
research questions. Of interest in the present endeavor 19 are two of her suggested research questions. First, what is the role of gender ideology in maintaining particular modes of production and the implications of these and what are the implications for the use of human built environ ments, such as cities? (for instance, why do many women still regard their domestic roles as both primary and "nat ural" and how does this constrain their use of urban space?); and the exploration of the close ties between the historically specific forms of different spheres of activi ty under capitalism (production, consumption, reproduction) and the acknowledgement that it is analytically incorrect to ignore these inter-relationships (Fincher, 1987:11).
1.4 ORGANIZATION OF THIS DISSERTATION
The body of this dissertation contains five chapters which document and attempt to conceptualize and explain the changing geography of female clerical work in the Columbus metropolitan area. Chapter 2 presents an overview of the literature explaining questions of social structure as they relate to the research question. The socialist feminist view of the interdependent nature of patriarchy and capi talism is outlined and employed in a discussion of the var ious forms of the gender division of labor in the twentieth century. As the relationship between patriarchy and capi talism is a dynamic one (as the massive post-Second World 20
War increase in the paid employment of women demonstrates) an attempt is made to explain the growth of female paid employment in general, and of female clerical work in par ticular .
Chapter 3 provides a discussion of the literature, exam ining questions of space which concern the growth and spa tial patterning of female clerical work. Specifically dis cussed are: the conception of spatial entrapment — an implicit theme in much of the previous work on the subur banization of female clerical work; the recent vintage of spatial entrapment as a social problem for women; and a review of the literature reporting on journey-to-work stud ies as one means of evaluating the validity of the notion of the spatial entrapment as it applies to suburban women.
The methods employed for the quantitative and qualita tive analyses are outlined in Chapter 4. This chapter also includes an elaboration of some of the theoretical issues already raised regarding social theory and geography.
The results of the quantitative analyses are reported in
Chapter 5. This includes an assessment of whether, in
fact, there has been a suburbanization of female clerical
labor in the Columbus metropolitan area. Also reported are
the results of analyses of the relationship between the
growth of clerical work in suburban locations and measures
of the need of married suburban mothers for paid employ 21 ment. Finally in Chapter 5 the results of a number of journey-to-work studies that have been undertaken are pre sented
Chapter 6 involves a qualitative analysis of some of the relationships that Chapter 5 reveals. This is achieved through an interpretation of unstructured, interactive interviews which were conducted with twenty-eight female clerical workers living in Westerville and Worthington; and with ten personnel managers of firms located in Westerville and Worthington which employ a large number of clerical workers.
The final chapter consists of a summary and conclusions drawn from the research presented. Some of the contribu tions of this dissertation to urban and social geography are also outlined. In sum, the body of this dissertation has three distinct parts: literature reviews (Chapters 2 and 3), methodology (Chapter 4), and analyses (Chapters 5 and 6). CHAPTER II
THE LITERATURES QUESTIONS OF SOCIAL STRUCTURE
2.1 INTRODUCTION
This Chapter reviews the literature exploring questions of social structure as they relate to the paid employment of women in general, and to women clerical workers in par ticular. The first section highlights the main theoretical premise of the socialist feminist view of the interdepen dent nature of patriarchy and capitalism. This is followed by a discussion of the various concrete forms of the gender division of labor in the twentieth century. It is argued that the relationship between patriarchy and capitalism is constantly changing, as the massive post-Second World War increase in the paid employment of women indicates. In an attempt to explain why this growth occurred, women's chang ing powers, wants and needs are outlined, these are concep tualized in terms of changes in the social structure. This is achieved by borrowing the concept of "Fordism" from the
French Regulation school. One way in which Fordism has had a clear impact on womens' paid employment is in terms of the rapid increase in the demand for clerical work. Fur-
- 22 - 23 thermore, it is argued that the recent increase in the paid employment of women, particularly those with young chil dren, is related to the collapse of Fordism, as, for instance, the recent surge in families needing the married woman's wages to maintain their standard of living.
2.2 PATRIARCHY AND CAPITALISM
This section outlines the socialist feminist view of the relationship between patriarchy and capitalism. Essential ly patriarchy and capitalism are conceptualized as two sep arate social relations of domination and subordination.
They do, however, interact. In particular capital has used pre-existing gender relations to assure its own production, just as men have used capitalist social relations to assure the reproduction of patriarchy. The interrelationship between patriarchy and capitalism is especially apparent in the gender division of labor. Capital has used gender relations in order to ensure its authority; that is, the relations of authority within the work-place are gender structured. At the same time men have used capitalism's monopoly of the means of subsistence to enforce their domi nation over women both in the market place (through unions
excluding women from wage labor, for example) as well as in
the domestic sphere. Thus this second section explores the
gender division of labor within the labor force. This will 24
include a discussion of how the gender division of labor underpins the conceptualization of "skilled work" and how
this impacts upon women. Attention will then focus on the
gender division of labor within the household.
2.2.1 The Debate
Recently socialist feminists appear to have reached some
agreement about the theoretical analysis of the oppression
of women under capitalisms conceptualization should proceed
as an interaction between patriarchal gender relations and
capitalist economic relations (Eisenstein, 1979; Hartmann,
1979; Mackintosh, 1977, 1981; McDonough and Harrison, 1978;
Quick, 1977; Vogel, 1983; Young, 1981). Some, such as
Eisenstein, even refer to "capitalist patriarchy" in an
attempt to:
emphasize the mutually reinforcing dialectical relationship between capitalist class structure and hierarchical sexual structuring. Understand ing this interdependence of capitalism and patri archy is essential to the socialist feminist political analysis. Although patriarchy (as male supremacy) existed before capitalism, and contin ues in post-capitalist societies, it is their present relationship that must be changed. In this sense socialist feminism moves beyond singu lar Marxist analysis and isolated radical femin ist theory (1979:5).
For socialist feminists this offers a viable alternative
to supposedly "gender-blind" marxist theory because, as
Hartmann (1979) points out in her seminal work "The Unhappy
Marriage of Marxism and Feminism", when a feminist and 25 marxist analysis are attempted simultaneously, the feminist account is often subordinated. Apparently all problems can be subsumed under capital, allowing the contribution of men as men to human exploitation and oppression to go un explored (Cockburn, 1984). As Cockburn explains:
Thinking in terms of two systems at once is not easy, especially perhaps to marxists who have such a well-developed sense of the class system From long practice, our eyes can distinguish a "mode of production" without difficulty. After all, we have had over a century of experience in teasing out this model from the mass of events we live through and read about (1984:194).
According to traditional marxist theory there is nothing inherent about the logic of capitalism which necessitates segregation of labor by gender (or race, or ethnicity) as
(theoretically) it is the first mode of production to be indifferent to the social identities of the people in the exploited class (making all workers equal in the market place). Indeed it is argued that there is (theoretically) an inherent tendency in capitalism to "dilute identities like gender or race as capital strives to absorb people into the labor market and reduce them to interchangeable units of labor abstracted from any specific identity"
(Meiksins Wood, 1988:5). In short, the development of cap italism can be interpreted as the development of "empty places" without regard for gender, race, or any other social identity. 26
However, as numerous scholars have pointed out, "pure" capitalism, indifferent to the specific types of people absorbed into the labor market and remorselessly extracting surplus value in its continuous cycle of accumulation, has never existed (Hartmann, 1979). Indeed when capitalism was first developing it encountered a pre-existing social structure, that of patriarchy (by which people were already stratified), which was then harnessed to capital's need for different kinds of labor (Hartmann, 1979; Phillips and Tay lor, 1980).3 As Mackintosh (1981:9) perceptively inquires:
"if women's subordination within society predates capital ism, then surely we cannot hope to explain it solely in terms of the inherent logic of the capitalist system?"
Thus a theory in which patriarchal gender relations are central is required in order to explain why it is usually women who are relegated to subordinate places within the capitalist hierarchy. Indeed, without the intervention of patriarchy, perhaps the predictions that capitalism would destroy gender divisions by indiscriminately absorbing
3 Indeed the whole notion of capitalism developing through the mechanical operation of certain laws of material development, has been attacked for tending towards eco nomic reductionism. In extreme interpretations of Marx not only has gender been ignored, but so have virtually all social factors, leaving only "the relentless grinding-through of an inner logic of capitalist develop ment, in which the imperatives of profitability push cap italist society along a pre-destined route ... an econom ic train carrying us all in one direction, regardless of who we are or the historical baggage we carry" (Phillips and Taylor, 1980:82). 27 women into wage labor would have reached fruition (Phillips and Taylor, 1980).4 Thus, as Burris (1982) points out in
4 Clearly the role of women in marxist theory is a conten tious one. Not only is marxist theory "gender-blind" with regard to paid workers it is also largely blind to the importance of women's unpaid domestic labor in a cap italist society. Humphries (1977) nicely summarizes this issue: "Marx abstracts from the problem of domestic labor by dealing with a situation in which all workers are engaged in capitalist production and perform no domestic labor whatsoever. No use-values are produced within the household and the capitalist sector provides everything required to replace the labor-power used up in production. This gives the reproduction cycle of labor- power a distinctive feature in that value is neither cre ated nor destroyed but merely recycled. V-Jages are used to purchase a subsistence bundle of commodities whose 'consumption' mysteriously leads to the replacement of used-up labor-power. This new labor-power is then exchanged for a new bundle of commodities which is in turn consumed, and so the process continues indefinitely, with labor power being used up in capitalist production and replaced through the act of consumption" (1977:143). Again "pure" capitalism of this form never existed. Instead the maintenance of capitalist society requires two kinds of work: production and reproduction. Thus women's role as an unpaid domestic laborer — maintaining and reproducing the paid labor force, keeping wages down and profits up (since these services .would have- to be paid for otherwise) -- is vital to a thorough understand ing of capitalism. As Young (1981:52) remarks: "Using the category of production or labor to designate only the making of concrete material objects in a modern factory has been one of the unnecessary tragedies of marxian theory". Thus the whole notion of dichotomous spheres; public/private (domestic), unpaid/paid, "unproduc tive "/productive, reproduction/production; is a falla cious one.
Hence a critical priority for marxist-feminists and socialist-feminists has been the reconceptualization of marxist analysis to include domestic labor, as the previ ous neglect of this vital component of capitalism means that traditional marxist analysis has been incomplete (Beechey, 1977, 1979; Humphries, 1977; Malos, 1980; Moly- neux, 1979; Seccombe, 1974). (This body of literature is generally referred to as the domestic labor debate which focuses around whether or not domestic labor is "produc tive" or not, and whether women's domestic labor subsi- 28 her review of Michele Barrett's (1980) works
even though it is possible to conceive of alter nate (non-patriarchal) forms of social arrange ments which are functionally compatible with the capitalist mode of production/ once certain forms of women's oppression are structurally embedded within capitalist relations of production they become inseparable from the ongoing reproduction of the mode of production in its present form. The articulation of patriarchal relations within the capitalist mode of production must therefore be conceptualized as the outcome and expression of structurally conditioned, but historically contingent, patterns of struggle which can never be reduced to a simple functional logic (1982:68-69).
A number of feminist scholars suggest that feminist analysis of capitalism should proceed by examining the development of the gender division of labor as vital to the structure of the relations of production. (Barrett, 1980;
Young; 1981). According to Barrett (1980:164-171) capital ist relations of production are based on two principles which differ from those shaping pre-capitalist production.
First, the separation of home and work, prompted by the development of large-scale industry and wage-labor systems.
Second, capitalism emerged as the first mode of production which did not require all potentially productive people to be employed, and which also necessitated a fluctuation in the proportion of the population which is employed. This enabled the construction of a labor force divided by the
dises capital or not. (Also see Coulson, Magas and Wain- wright, 1975; Della Costa, 1975; Finch, 1983; Gardiner, 1975; Humphries, 1977; Hunt, 1980; Mackintosh, 1979; Wal- by, 1985).) 29 needs of capital to increase "economic productivity" and to control labor. Both principles are essential to capital and both have been manifested as forms of oppression of womens women have primary responsibility for the domestic sphere and women are in subordinated positions in the paid employment hierarchy. For Young (1981), as for Barrett, the "marginalization of women and thereby (their) function ing as a secondary labor force is an essential and funda mental characteristic of capitalism" itself (1981:58).5
Reviewing Heleieth Saffioti's work, Young argues that given that capitalism was the first mode of production not requiring all potentially productive people to be employed, and that it necessitated a fluctuation in the proportion of the population employed, some criterion was needed in order to distinguish between core workers and marginal workers.
Given an initial gender differential and a pre-existing sexist ideology, a patriarchal capitalism in which women function as a secondary or marginal labor force is the only historical possibility.6
5 By "essential and fundamental" Young does not mean that a capitalism without the marginalization of women is not logically possible (indeed as pointed out earlier, it is logically possible).
6 Young also points out that capitalism can also divide the labor force by race and ethnicity, when such criteria are present, but feels that gender is the most obvious and permanent division. 30
2.2.2 Gender Division of Labor within the Labor Force
Although there are few (if any) jobs which, by their content, necessitate only one gender to do them, many jobs have become largely associated with either women or men, and, as a result, have become "sex-typed" as either
"women’s jobs" or "men's jobs" (Dex, 1985; Fox and Hesse-
Biber, 1984).7 In other words there exists a clear hori zontal gender division of labor within the paid labor force, with most women and men employed in different sec tors of the economy.
Table 5 shows that over the course of this century women have become more heavily concentrated into fewer occupa tions. To illustrate, women have come to be disproportion- ally represented in the "pink-collar" occupations such as clerical (35.2%) and service (19.5%) work, and underrepre sented in blue-collar occupations such as craft-work
(1.8%), operatives (10.7%), and laborers (2.1%). On the other hand, men are more evenly spread throughout the cat egories and are more likely to be managers, blue-collar crafts-workers, laborers and farmers (Blau and Jusenius,
1976; Bruegal, 1979; Milkman, 1976; Treiman and Terrells,
1975). That half of all women in paid employment are
7 "Women's jobs" have come to include typists, secretaries, office cleaners, hairdressers, sewing machinists, wait resses, and maids (jobs in which over 70% of the workers are female); and "men's jobs" include mechanics, trans portation workers, farmers, miners, crafts-workers, architects, and dentists (jobs in which over 70% of the workers are men). 31
Table 5
Percent of all Women in Paid Labor in each Occupation Class
OCCUPATIONAL CLASS 1910 1930 1950 1970 1980 Professional, excluding nurses and non-college teachers 2.2 4.1 4.1 5.9 8.3 Nurses and non-college teachers 6.9 10.1 8.3 9.4 8.5 Proprietors, managers and administrators (incl. farm) 6.0 5.4 5.0 3.8 7.3 Clerical and kindred 7.3 18.5 27.2 34.4 35.2 Sales workers 5.7 8.4 8.4 7.4 6.8 Craft and kindred 7.6 2.3 1.5 1.8 1.8- Operatives 15.2 16.5 19.5 15.0 10. 7 Laborers 18.6 6.5 3.7 1.6 2.1 Services Household 16.3 15.2 8.6 3.9 2.5 Non-household 14.3 13.3 12.2 16.6 17.0 All Occupations 100.0 100.0 100.0 100.0 100.0
Source: Mallier and Rosser (1987:48)
located in just twenty-one occupations, whereas half of men may be found in a total of sixty-five occupations succinct ly illustrates the gender division of labor within the paid labor force (Coser, 1982). Moreover one quarter of all women who work are employed in just five occupations — secretary, book-keeper, household worker, elementary educa tion teacher and waitress.8
8 Accordingly Kassell (1984) estimates that nine out of ten people have co-workers of the same sex. On the whole most women and men work in occupations with gender ratios of at least four to one. It should be noted that the extent of segregation has declined somewhat; in 1970 60% of women and 75% of men worked in occupations where the gender ratio is at least four to one; whereas in 1980 the equivalent figures were 46% and 53%. Thus very few peo ple (15% of all women and 11% of all men) work in truly integrated occupations (Rytina and Bianchi, 1984), and 69% of working men would have to switch jobs in order to eliminate the gender division of labor within the United State's labor force. 32
Table 6 indicates that while in the aggregate women have become better represented in all occupational classes (rep resenting 21% in 1910 and 41.9% in 1980) this disguises certain important variations. For instance, although the proportion of all professionals and managers who are women has increased, the proportion of blue-collar workers who are female has declined. Perhaps most significantly in the present context is the extensive feminization of pink- collar occupations, especially clerical and sales work.
Women made up 34.3% of clerical and 14.5% of sales workers in 1910; by 1980 these figures had risen to 80.1% and 45.3% respectively.9 In short, and as Sapiro (1986) puts it:
Women clean, cook and serve food, sew, teach, and do clerical and sales work; in other words, much of the work women do for pay is the same kind of work they are expected to do as wives and mothers at home (1986: 384-85).
Much research and media attention has focused on those women moving into non-traditional (ie. "men's") occupa tions, the implication being that the labor force is becom ing less segregated. However as Tables 5 and 6 clearly indicate, this is not really the case. Certainly there are now more women in the professions than ever before, but these women represent a relative minority of women workers.
In short the reality is that the occupations of most women
in the 1980's are not that different from those of women in
9 Also note that services, already highly feminized in 1910, became even more so by 1980. 33
Table 6
Percent of Workers who are Women in each Occupational Class
OCCUPATIONAL CLASS 1910 1930 1950 1970 1980 Professional, excluding nurses and non-college teachers 17.5 23.0 19.7 22.7 30. 7 Nurses and non-college teachers 81.4 80.3 81.5 76. 7 78.3 Proprietors, managers and Administrators (incl. farm) 6.7 7.6 8.6 14.6 24.2 Clerical and kindred 34.3 49.4 62.2 73.8 80.1 Sales workers 14.5 18.3 34.0 40.1 45.3 Craft and kindred 11.5 3.6 3.1 4.9 6.0 Operatives 22.5 23.2 24.0 31.8 32.0 Laborers 17.9 11.5 10.1 10.2 14.2 Services Household 83.3 81.8 94.9 96. 9 97.5 Non-household 46.8 50.8 44.9 55. 8 58.9 All Occupations 21.2 22.0 27.8 37.2 41.9
Source: Mallier and Rosser (1987:54) the paid labor force in the 1950's.10
In addition to the horizontal gender division of labor, there is a vertical gender division of labor. Surveys of employment status show that women occupy the lowest status, lowest skilled jobs that require the fewest qualifications and training (Roos and Reskin, 1984; Women and Geography
Study Group of the IBG, 1984). So, even though 70% of office workers in Britain are women, only 14% of office managers are female (Lewis, 1984), while in the U.S. only
25% of all managers and administrators (1% of managers
earning over $30,000) are women; and although seven out of
ten bank employees are women, only one will ever obtain a
10 Furthermore, as we shall see, the bulk of the increase in women's labor force participation has been absorbed by the fastest growing sectors of the economy which may be characterized as offering largely "women's jobs". 34 managerial position (Nussbaum, 1982).11
Hence women, to the extent that they are employed, usu ally get less pay than men. This, along with the gender division of labor, is underpinned by a variety of widely held meanings regarding skills or lack thereof, the sex typing of jobs, women's work, woman's domestic role and
"pin money". In 1983, women working full-time, year-round, earned 64 cents for every dollar that a man earned (Smith and Ward, 1984), and only ten percent of women earn more than the median income (approximately $20,000) of men
(Pear, 1983). This rate has remained remarkably constant since the 1950's despite legislative and legal changes.12
As Table 7 shows, even in the same occupation, including those in which women predominate, women earn less than men
(Rytina, 1982).
11 Generally speaking "female" occupations have limited opportunities for on-the-job training or upward mobility within that occupation, but the same does not apply to male occupations. For example, while engineering and finance are considered acceptable stepping stones to management, elementary school teaching and nursing do not tend to lead to school or hospital administration positions (Larwood and Gutek, 1985).
12 This topic in itself has stimulated much research in the last fifteen years (see for example: Blau, 1984; Cor coran and Duncan; 1978; Fuchs, 1971; Treiman and Ter rell, 1975; Parcel and Mueller, 1983) but will not be covered in any detail in this dissertation. Suffice to say that generally it is argued that the gender wage gap can be explained only partially by women's lower job tenure, level of education, and the segregation of women into low paying jobs. The "remaining unexplained dif ferential" is generally attributed to unquantifiable factors such as gender discrimination and the exploita tion of women as cheap labor (Zorn, 1985). 35
Table 7
Median Earnings of Women and Men by Occupation, 1983,
Earnings of women as a percent of earnings Occupations men women of men
Executive, Administrative, and Managerial $30,476 $18,277 60.0 Professional 29,547 19,202 65.0 Technical and Related 24,573 16,555 67.4 Sales 23,128 11,979 51.8 Administrative Support, including Clerical 20,833 13,473 64. 7 Service 14,688 9,228 62.8 Precision Production, Craft, and Repair 21,520 13,245 61.5 Operators, Fabricaters, and Laborer 18,085 11,371 62.9
Note: Figures for workers employed full-time, year round.
Source: Blau and Ferber (1986:175)
Furthermore, that women get paid less than men is under pinned by a variety of widely held meanings regarding, for example, occupational segregation by gender, women's domes tic role, and skilled work. Gender roles, relations, and ideology saturate the meaning of "skilled work". Often much of what is considered "women's work" is dubbed "unskilled" simply because it is women who do it (Lewis, 1985; Phillips and Taylor, 1980). "Skill" is a socially constructed con cept, and as Phillips and Taylor (1980) argue:
Women workers carry into the work-place their status as subordinate individuals, and this sta tus comes to define the value of the work that they do. Far from being an objective economic fact, skill is often an ideological category 36
imposed on certain types of work by virtue of the sex and power of the workers who perform it (1980;79).
Numerous empirical studies illustrate this argument and some even highlight the contradiction therein. In studies of skill distinction in the production of paper boxes in
Britain, women who worked on hand-fed machines were cat egorized and paid as unskilled workers; whereas men produc ing cartons on more automated machinery (requiring less individual concentration) were categorized as skilled or semi-skilled and, of course, paid accordingly (Craig,
Rubery, Tarling, and Wilkinson, 1982; Rubery and Wilkinson,
1979). Thus the "women producing paper-boxes are simply women producing paper-boxes, and however much the work itself might seem to qualify for upgrading, it remains uns killed because it is done by typically unskilled workers — women" (Phillips and Taylor, 1980:84). Similar conclusions result from comparable studies such as Birnbaum's (quoted in Phillips and Taylor, 1980) and Coyle's (1982) studies of the clothing industry.13 In addition even when skill lev els or sectors are re-classified, slight differences between jobs are often employed in such a way that the skilled workers still end up being men and the unskilled workers women.
13 These two studies, among others, point to the role of craft-based unions and apprenticeships in retaining men's craft domination at the expense of women. 37
This process takes a slightly different form in the non manufacturing industries. Significantly it has been the non-manufacturing (and thus largely non-union) sectors of the economy which have expanded since the Second World War, providing an increasingly larger proportion of jobs for women. A new form of labor, low skilled and certainly low paid, could be explicitly built on the notion of women's paid work as unskilled or semi-skilled. Clerical work pro vides such an example. A new kind of labor was created, defined as "female" from its inception, with little or no comparability with pre-existing men's jobs. The transfor mation of clerical labor from the male "black-coated" work er of the nineteenth century to the female office temps, file clerks and typists of the 1980's occurred through the huge expansion of offices after the First World War (Cromp ton and Jones, 1984; (Crompton, Jones and Reid, 1982; Dav ies, 1975, 1982; Davy, 1986; McNally, 1979) As Phillips and
Taylor describe:
The new class of clerical workers had little in common with the clerks of the previous century, and the skill component of their work was immedi ately downgraded to typically "female" abilities -- including, as usual, dexterity, ability to carry out repetitive tasks, and so on. In this case it was not that men's jobs were deskilled, and women drawn into them, but that a new catego ry of work was created which was classified as "inferior" not simply by virtue of the skills required for it but by virtue of the "inferior" status of the women who came to perform it (1980:84). 38
Phillips and Taylor (1980) go on to intimate that this pro cess of degrading would have involved a good deal more class struggle had it been men rather than young women who were being employed. They also note that the personal ser vice aspect of clerical work, which is often a central com ponent of many women clerical workers employment, would not have developed as it has if it were men rather than women who had filled the newly expanding office positions.
2 .2.3 Gender Division of Labor within the Household
Prior to the rise of industrial capitalism women played a very active role in the economic life of society: domes tic and "economically productive" labor were fused into one way of life in the home. The rise of capitalism saw the separation of these "spheres", as "economically productive" labor increasingly took place in the large-scale factory under the wage labor system. While a gender division of labor had clearly existed before capitalism, in many cases gender roles had overlapped and there had been a degree of continuity between many of women's and men's tasks.
During the initial stages of capitalism women (and chil dren) worked alongside men, and they only began to withdraw as wages rose and when, in the nineteenth century, wives
and mothers were expected to stay at home, like the women
in more privileged social groups (Hartmann, 1976; Oakley, 39
1974).14 The male domination of the pre-industrial family unit was not only carried over into the factory; it was also fortified through craft-based union (male) exclusive ness, apprenticeships, and men’s power to sub-contract work
(Lewis, 1985). As Young remarks:
While women in pre-capitalist society were by no means the social equal of men, all evidence points to the conclusion that our situation dete riorated with the development of capitalism. In pre-capitalist societies women dominated a number of crucial skills, and thus their labor and their knowledge were indispensable to the family, the manor, and the village. In many craft guilds of the sixteenth and seventeenth centuries women were members on equal terms with men, and even dominated some of them. Women engaged in indus try and trade. Pre-capitalist culture understood marriage as an economic partnership; men did not "support” women. The law reflected this relative equality of women by allowing them to make con tracts in their own name and retain their own property even in marriage.
By the nineteenth century women's economic independence had been almost entirely undermined and her legal rights were non-existent. Capital thrust women for the first time in history to the margins of economic activity (1981:59).
So as capitalism advanced, the development of factory pro duction and wage labor outside the household served to intensify the pre-existing gender division of labor. The value of labor came to be assessed in monetary terms, changing the meanings people attached to "work". Labor within the "public sphere" became "real" work while domes
14 This is somewhat of an over-generalization as many poor, working-class, immigrant and Black women have always worked for pay outside the home (Kessler-Harris, 1981, 1982; Tilly and Scott, 1978). 40 tic labor, although essential for the reproduction of the labor force, came to be considered of little economic or social value. Accordingly the domestic (women's) and pub lic (men's) spheres became increasingly more distinct, changing the very meaning of "home". As Matthaei (1982) puts it:
When the development of industrial capitalism separated commodity production from the house hold, the family was freed from the function of organizing this production, and it was freed from the presence of strangers in the family. The household became a home, a private family place. Family relationships -- between husband and wife, between parents and children -- began to gain a content of their own. The sharing that had char acterized the household developed an emotional underpinning — love and personal attachment between family members -- when the household became a home. This new type of social relation ship was a necessary complement to the new free dom and competition in the economy (1982:110).
Thus the forms that the household and the family take today are different from the many forms it has taken in the past. Indeed today's so-called "traditional family" is, in reality, a fairly recent form of the family which emerged during the late nineteenth century and captured popular imagination as the "ideal" family form. Among other things, this "bourgeois family ideal" was fostered by the gender ideology of the "cult of true womanhood". This reinforced the notion that women were frail and delicate, uniquely endowed with the emotional qualities required to maintain the domestic sphere and protect society's moral L 41 fabric from the corrupting influences of capitalism.15
Further, mothers became "the obvious source of every thing that would save or damn the child; the historical and spiritual destiny of America lay in her hands" (Wishy,
1972), and soon this feminine ideal developed into a belief that a nurturing full-time mother was necessary for the healthy functioning of American families and should be every woman's priority in life (Ryan, 1979; Welten, 1966).
As Glazer (1980) clearly illustrates, motherhood has become increasingly more complex:
In the nineteenth century, mother's work appeared at first to be the relatively simple tasks of seeing that children learned certain moral stan dards, were fed, were kept clean, and slept regu lar hours. Gradually, broader and more difficult work was assigned to mothers. Contemporary moth ers (those of the last 100 years for the working- class and the last 150, perhaps, for the bour geoisie) have had to build character. This century, mothers have gradually come to be responsible for the building of personality, too. They must develop capabilities in their children during early childhood that will be compatible with demands of the school, the work-place, and political life ... Mothers are responsible for the "happiness" of their children as well as their general fitness for life in a capitalist society (1980: 260-261).
Initially this gender ideology only informed the prac tices of the bourgeoisie (it certainly did not apply to the slave-owners of Black women nor the employers of the first
15 Such a gender ideology only made sense in a mode of pro duction which does not require all potentially "produc tive" members of society to be employed, but which requires a fluctuation in the actual proportion of the population employed. 42
European immigrant women), but post-Second World War eco nomic prosperity enabled many working-class families to attain this bourgeois ideal. Real incomes rose, single
family housing was relatively affordable and plentiful
(especially in the suburbs), consumer durables became more available, and there was an increase in the number of wife- husband (as opposed to extended family) households. Young adults were increasingly able to marry, buy houses and
establish residences independently of their parents (see
Section 2.3.2). This, along with the baby boom, saw the
rise of the full-time "house-wife" who was a "female angel
of the hearth", often cloistered in a suburban home, dedi
cated to the reproduction of her family and the pursuit of
an arcadian life-style where "home is a haven" from the
ravages and brutal reality of the work-place.16 Indeed the
gradual spatial separation of production to the city and
reproduction to the suburbs has been seen by some as pro
viding some of the basis for the development of the social
ly defined "house-wife" (Bowlby and McDowell, n.d; Davi-
doff, 1977; Mackenzie and Rose, 1983; Lasch, 1977, Saegart,
16 Gender ideology was further entrenched by the rise of professional "experts" (especially doctors and psycholo gists) emphasizing the need to "re-build" the family in the post-war era. Accordingly the mother/child relation ship was seen to be the key to the healthy development of the post-war generation (Lewis, 1985; Riley, 1983). Child-care was modified by Fruedian-inspired theories of child development. Mothers were (as they continue to be) cautioned to care for the child's social and mental development, as well as their health, discipline and cleanliness (Vanek, 1974). 43
1981).
In the 1950's, therefore, the gender division of labor within the household was clear cut: men were "bread winners" and women did the domestic chores and children- rearing. In the 1980's, however, over half of all women are in paid employment, many consider their jobs to be careers, and contribute an increasingly larger share to their family's income. Yet despite the social changes which have enhanced women's ability and need for wage work since the 1950's, the household gender division of labor does not appear to have altered accordingly. Studies sug gest that even when a wife is in paid employment, she still carries her double load with little assistance from her spouse (Berk, 1980, 1985; Berk and Berk, 1979; Blau and
Ferber, 1986, Ch. 2; Sharpe, 1984; Vanek, 1974, 1980).
Generally it is still women who do the domestic duties
(such as housework, grocery shopping, and child-care, including taking time off work when they are sick), although it does appear to be becoming more equitable among couples in their twenties and thirties, especially when it comes to child-care17 (Berk, 1980, 1985; Berk and Berk,
17 Whether their wives are in paid employment or not, mar ried men do an average of four to thirteen hours of domestic chores a week. (Meissner, Humphreys, Meis, and Schein, 1975; Pleck, 1983; Robinson, 1977; Walker and Woods, 1976, cited in Blau and Ferber, 1986). In addi tion, although there is only limited supporting evi dence, it seems feasible that older children of employed mothers play an important role in the household gender division of labor (Blau and Ferber, 1986). 44
1979; Blau and Ferber, 1986; Sharpe, 1984). In short, paid employment has not radically altered the basic gender role of women as "natural" caretakers of the home and family.
Indeed Bose (1982) suggests that household technology and appliances are used as a substitute for a more equitable household gender division of labor and Sharpe (1984) illus trates this by suggesting that:
Many men would doubtless prefer to be able to equip their wives with a super-technology kitchen than share housework themselves, but this in fact does not necessarily reduce the amount of time women spend there. How much time and labour is saved by a (dish washer), for instance, when you have first to clean off the food, stack the items in specified places, wash burnt pans by hand, and then unload it all again later? (1984:175)
Thus the ideology that dictated that a woman's primary role was that of care-giver in the home still persists despite a changing reality. So the form of the gender division of labor within the household and even in the labor force is still largely based on the assumption that the "tradition al" family form remains still the predominate one. For instance, marriage partners who are both in the paid labor force, often relate to each other along traditional patriarchal lines, which in turn affects the allocation of household resources in such a way that the husband usually has more access to the family car leaving the woman to rely on slower, less flexible public transportation (Dix, Car penter, Clarke, Pollard, and Spencer, 1983; Pickup, 1984). 45
While in the paid labor force mothers face massive obsta cles to bringing the fathers in as full partners in the parenting process, as few men can withdraw from the labor force to care for their new-borns without risking consider able penalties. And as the man's salary tends to be great er than the woman's, few couples can afford to take advan tage of such opportunities even when they do exist (Gerson,
1983).
2.3. PATRIARCHY AND CAPITALISM IN HISTORICAL PERSPECTIVE
The concrete outcome of the interaction of patriarchy and capitalism has not remained stable over time. During the course of the twentieth century there have clearly been changes in gender ideology and there has been some change in gender relations, as indeed the increase in the numbers of women in the paid labor force indicates. These changes can be explained in terms of women's changing powers, wants and needs. So, for instance smaller families and the
introduction of household appliances, for example, have arguably decreased the amount of time and labor involved in women's domestic roles. The increased numbers of divorced women and female householders mean that more and more women need to enter the paid labor force. At the same time,
societal changes such as the increased emphasis on equal opportunities and education achievement have raised women's 46 aspirations and expectations (especially white, middle- class women). In addition the Women's Movement and legis lation regarding women's work have played roles in altering both the attitudes, and to some extent, the reality of working women's livfes. Women have these changing powers, wants and needs, however by virtue of social structural change. One approach to conceptualizing that change which is being used increasingly in the social sciences is that of the French Regulation School which will be discussed in section 2.4. This section will then examine some of the factors which have contributed to the transformation of women's roles, particularly since the Second World War, increasing both the ability and need of women to take on paid employment.
2.3.1 The Changing Character of Motherhood
The birth rate has been declining since before 1900, when women were having an average of four children.
Excluding the baby boom years, this decline has continued throughout the twentieth century and by 1980 child-bearing had fallen below the natural replacement level (Sternlieb,
Hughes, and Hughes, 1982). In fact there has been an
increase in the preference among wives for two-child fami
lies from 25% in 1960 to 50% in 1983, and a concomitant decline in the preference for larger families: in 1960 37% of wives wanted four children and in 1983 this figure had 47 dropped to 7% (Bianchi and Spain, 1986). There has also been a rise in the rate of childlessness or one child fami lies (Masnick and Bane, 1980). For instance, in 1983 62% of currently married women aged 30 to 34 with one child expected to have no more children (Bianchi and Spain,
1986). Some argue that there has been an increase in per manent childlessness: although not yet known for women presently in their twenties and thirties, it is estimated that 25% of White women and 20% of Black women will remain childless (Bloom, 1982; Bloom and Trussell, 1984).
In addition to the decline in family size there has been, since the Depression, an increase in the proportion of women delaying child-birth (Bloom and Trussell, 1984;
Hofferth, 1984). Among women born between 1935 and 1939
(mothers of baby-boomers), 71.6% had had their first child by the age of twenty-five. Among those born between 1950 and 1954 (baby-boomers), this number had fallen to 61.2%.
At the same time, looking at women who have already reached menopause, the median age of women on birth of their last child has been declining since the last century (Bianchi and Spain, 1986). Thus women's child-bearing years are being compressed into a shorter period of time by the planned, close spacing of children (Westoff and Ryder,
1977) . 1 8
18 Reliable contraception has made it easier for women to control their fertility, although the issue of whether women should have control over their fertility has long 48
If one adds to this declining fecundity increasing lon gevity, the result is a decrease in the percent of women’s
lives committed to motherhood over the last seventy years
(Klein, 1984; Tilly and Scott, 1978). Not only has the pro portion of a woman’s life spent caring for young children
decreased, but the percent of her life beyond the point at
which the children leave home has increased. This makes it
possible for women to spend more of their life-time in the
labor force.19
been a major political, social and religious issue (Gor don, 1977). The most effective forms of birth-control -- the "Pill", IUD and sterilization -- were not widely available until the 1960's and 1970's. Abortion was legalized in the United States in 1973 (Roe vs Wade). The ratio of abortions to live births increased from 22:1000 in the early seventies to 36:1000 in 1980, and the vast majority of women (65% in 1980) seeking abor tions are under the age of twenty-five.
19 Accordingly it appears that women who work for pay out side the home have fewer children than those who do not. In 1983, women in the paid labor force aged 18 to 34 had an average of 0.8 children, while those women not employed had an average of 1.7 children. Furthermore thirteen percent of employed women compared with six percent of those not in the paid labor force expect to remain childless (Bianchi and Spain, 1986). There is some controversy over the direction of the "causal arrow" here. Some argue that lower fertility rates have increased women's labor force participation, suggesting that the ages and number of children affect the prob ability of a woman being in the paid labor force as well as whether she works full- or part-time (although the same does not hold true for men) (Ericksen, 1977; Fagna- ni, 1983; Hanson and Hanson, 1980, 1981; Smith, 1979; U.S. Department of Labor, 1980). While others argue that women plan to reduce their fertility in order to attain their labor force goals (Waite and Stolzenberg, 1976; Stolzenberg and Waite, 1977). So, for instance, one study showed that women's plans to be in the labor force when they are thirty-five years old had a substan tial effect on the number of children that they expected 49
2.3.2 Women in "Non-Traditional" Households
The divorce rate in the United States has increased from
0.9 per 1000 persons in 1910 to 2.6 in 1950 to 5.1 in 1982
(Blau and Ferber, 1986). Among marriages begun in the
1950's approximately thirty percent have or will end in divorce; for those marriages begun in 1970 it is estimated that this figure will rise to an all-time high of about 50%
(Cherlin, 1981). Indeed the proportion of women over age fifteen who reported themselves as divorced rose from 2.9% in 1960 to 7.1% in 1980.20 Over the same period the per cent of never-married 20- to 24- year olds almost doubled from 28% to 53%. For 25- to 29- year old women the percent has more than doubled, and it almost doubled for women aged
30 to 34. There was a similar pattern for men (England and
Farkas, 1986). So as Gerson (1983) points out:
marriage no longer offers the promise of perma nence upon which female domesticity depended in the past. Increasingly marital instability means that women can no longer equate marital vows with economic security. In this context, work takes on greater importance in women's lives and at the same time motherhood recedes as a central concern or all-consuming occupation (Gerson, 1983:143-144).
to have (Jones, 1981).
20 The number of couples separated is actually higher than the number of divorced, because some couples separate but never divorce. In addition, many couples live together, never marry and then separate (Cherlin, 1981; Kingsley Davis, 1983). And the number of co-habiting couples has tripled since 1960 to 1.9 million in 1980 (Bianchi and Spain, 1986). 50
Thus for many women paid employment is necessary because
they are the sole income earners of their respective house holds, and because divorce has an unequal economic impact by gender. Estimates suggest that following a divorce or
separation the financial status of women will decrease by
about seven percent, while men's financial status will
increase by about seventeen percent (Cherlin, 1981;).21
Since the Second World War the number of households in
the United States has been growing at a faster rate than
population growth. At the same time the average household
size has fallen from 3.37 persons per household in 1950 to
2.75 persons per household in 1980, so households are
becoming smaller (Sternlieb, Hughes and Hughes, 1982).
21 Women usually get custody of their children after a divorce and so women become single mothers, whereas men just become single. Hence, women tend to maintain a larger household than their former husbands. However most divorced women are awarded meager amounts of child support, the 1981 average being about $2206 per year per family not child (U.S. Bureau of Census, 1981). More over less than half of these women receive the full amount awarded, a quarter receive partial payment and a quarter receive nothing. Furthermore the chances of even being awarded child-support in the first place vary by race: Blacks have only a 29% chance compared with 71% for White women. All of this means that the per capita income of the home of the average child is declining (Pearce, 1983; Ohio Senate Task Force on Women Single Heads of Households, 1985). So as men tend not to meet either their alimony (which is often not awarded anyway) or child-support payments they have more disposable income on top of their already superior economic posi tion over their former wives. 51
Table 8
Changing Composition of Households; 1940 to 1980
HOUSEHOLD TYPE 1940 1950 1960 1970 1980
Total Households 35,087 42,251 52,809 63,573 80,389
Family Households 90.0% 88.1% 85.0% 80. 5% 73.3% Married Couple 75.8 non-working wife — 59.4 51.2 41.6 30.3 working wife -- 19.6 23.2 28.9 30.6 Female Householder 9.9 8.0 7.8 8.7 10.5 Male Householder 4.3 3.1 2.4 2.6 2.6
Nonfamily Households 10.0% 11.9 15.0 20.0 26.8 Female Householder 5.4 7.1 9.4 12.3 15.3 Male Householder 4.6 4.4 5.6 7.3 11.4
Sources Bianchi and Spain (1986:88) and Gerson (1982:140)
Today the "traditional family" (married couple, full-time home-maker wife and children) is no longer the predominant family form, accurately describing a mere 13% of American households (Diamond, 1982). As Table 8 clearly shows, it now co-exists with fast growing numbers of households head ed by women, single and childless adults, and two-earner couples. The married couple/non-working wife household has declined over the period while most other forms have risen.
The greatest increase has been among married couples with working wives and among nonfamily householders. A signifi cant component of this recent change has been the increase
in the overall number of households headed by women, up from 15.3% in 1940 to 25.9% in 1980 (compared with a 8.9% to 14% change for male householders over the same period). 52
In part this is due to the increase in divorce and women out-living their partners, but it is also due to women delaying or shunning marriage altogether and women forming households separately from their parents. Clearly these changes have had an effect on family structure, particular ly in terms of the increase in one parent families, often headed by a woman working to support her family. Further more there is a greater probability of children in a house hold headed by a woman (60% in 1983) than in one headed by a man (37% in 1983). Indeed female householders are repre sentative of the increasingly altered nature of parenthood and the situation of children. In 1960 88.5% of children lived with both their parents; by 1980 this figure had fallen to 76.6% (Bianchi and Spain, 1986).22
22 More than fifty percent of women who head households earn less than $10,000 per year (compared with only thirteen percent of "traditional" households) and more than half of the families living in poverty are main tained by women. The percent of families below the pov erty line headed by women was 21.3% in 1960, and increased to 43.8% by 1980. Accordingly 52.7% of all impoverished families with dependent children were female headed households, up from 43% in 1970. Thus 56% of children aged less than nineteen who lived in a female headed household in 1982 were in poverty, com pared with 13% of children in other families (Preston, 1984). It should be noted that these figures vary by race as in 1982 57% of persons living in Black house holds with a woman householder were in poverty, compared with only 29% of similar White households (Bianchi and Spain, 1986 *.208). 53
2.3.3 The Redefining of Children
Historically families have not been particularly child- centered. Up until the early twentieth century,most chil dren were expected to be facilitaters of their family's budget, particularly in working-class families given their susceptibility to periodic unemployment and illness (Hare- ven, 1977; Hogan, 1978). To illustrate, in 1884 the Illi nois Bureau of Labor Statistics conducted an investigation of the State's working-class families. They found that in cases where children worked they earned 36% of the total income of their families. Indeed it was found that a stag gering 80% of the state's working class families relied upon the earnings of their children in order to survive.
In other words in many nineteenth century families it was economically necessary for children to work (Hogan, 1978).
Indeed many children left home, resided at boarding houses and worked for pay well before what would be considered an appropriate age for such activities today (Zelizer, 1985).
Children continued to be a source of wage labor into the twentieth century. For example, in 1900 by the age of twelve a sizable proportion of children had left school in order to secure paid jobs (and most gave their income to their families). Strict child labor laws were not really enforced until the 1930's (Glazer, 1980). Indeed the whole notion of childhood and adolescence as separate from adult 54 hood is a historically recent social construction tied to the rise of compulsory education,, child labor laws and a capitalist industrial economy (Cox and England, 1988).23
Education has been mandatory in the United States since the early twentieth century. Generally speaking, the his tory- of education has been punctuated by major class strug gles over the introduction of, and length of, mandatory schooling. The middle-class positively encouraged the institutionalization of the education process, but such legislation found little favor among working-class parents.
Given their reliance on their children's wages to bolster family income, it is not suprising that working-class parents opposed mandatory schooling and that truancy was rife among working-class children.24
23 In addition, in the absence of the welfare state chil dren's earnings were also an investment for parents old age.
24 The bourgeois response is well documented. Issel (1979), for example, quotes a contributor to the Penn sylvania School Journal of 1881 as declaring that: "The The increase of foreign immigration, the increase of the Negro population, and the disposition of parents in man ufacturing cities to send their children to work rather than to school ... will eventually work great injury to the nation". Indeed such retorts reflect, according to Katz (1977), a common concern among nineteenth century education reformers. There was a fear that the influx of predominantly rural immigrants in to the States was generating social disorder. Education, or more specifi cally, 'Americanization', was regarded to be the pana cea. Furthermore, as Katz has pointed out, a captive audience could more easily be inculcated with the capi talist industrial norms pertaining to "proper" behavior. As such, the mechanisms that would lead to nation-wide compulsory education were set in motion (England, 1985). 55
Thus today when children leave high school they are no longer sent out to work in order to supplement the family budget. They are more likely to be going to college or forming their own households (Pahl, 1984). It could be argued that as families can no longer rely on their older children to work as they have previously, it is now women who are supplementing the family income. Indeed many moth ers find that they must go out to work in order to pay for their children's education. Moreover, from being economi cally useful a child now gains from being economically use less, but emotionally priceless. No longer a facilitater of the family budget, the child is now a vehicle for her/ his parents to realize their emotional needs. Indeed chil dren have become a luxury, given that they cost approxi mately $100,000 to $140,000 to raise (Zelizer, 1985).
2.3.4 Household Technology
The process of industrialization ensured that domestic work carried out by women became less and less concerned with the making of goods for home consumption, and more and more concerned with using goods manufactured elsewhere to ser vice other household members. For example, a typical North American married woman in 1800 would have lived on a farm, made her family's clothes, soap and candles, baked her own bread, looked after the kitchen garden and raised ani mals and poultry for food. By 1950 she would have lived in a city, bought manufactured clothes and cleaning materials, used electric light and bought canned, packaged and frozen food (Women and Geography Study Group of the IBG, 1984: 43). 56
Thus many domestic tasks became modified and more consumer durables became affordable to larger numbers of people.
Increasingly items previously considered luxuries have become necessities (such as a refrigerator), either through advertising25 or through the erosion of the original knowl edge or means of producing them (Braverman, 1974; Ehrenr- eich and English, 1975). Indeed Braverman (1974:281) notes that people have become increasingly "enmeshed in a web made up of commodity goods and commodity services from which there is little possibility of escape". While Pahl and Wallace (1985:223) remark that "the more capital goods and equipment (people) own, the more they are dependent on market services to maintain them".
In general, household technology (largely introduced in the last 150 years) has allowed many household activities to be displaced and mediated by consumer goods and services and, although this is a moot point, enabled women to take on waged labor (Bowlby and McDowell, n.d; Davidoff, 1977;
25 Indeed it is argued that the advertising industry is closely related to the increase in the commodification of goods and services in the earlier part of the century (Cowen, 1974; Ewen, 1976). Adverts reflected the grow ing awareness of women's buying power with regard to domestic products. As McDowell (1983) notes: "women who were freed from domestic drudgery by appliances, who were healthy, reasonably educated, and relatively afflu ent, were 'free' to choose clothes, automobiles, and endless domestic appliances. The boast that women wielded 75% of the purchasing power in the USA was not taken lightly. 'Professionalisation' and specialization of domestic products followed: women were presented by advertisers as experts choosing between different prod ucts in order to run an efficient home" (1983:66). 57
MacKenzie and Rose, 1983; Miller, 1983; Robinson, 1980;
Strasser, 1980). Historically, improvements in household technology may be labor-saving or time-saving or both.
Some of the most important improvements have included switches from coal stoves to gas or electric stoves, home made to mass produced clothing, as well as innovations in food processing, preserving and preparation, and lighting, heating and sanitation (Armstrong and Armstrong, 1983; Pow er, 1983).26 More recently freezers, washing machines, and micro-waves (as well as fast food restaurants and pre packaged meals) have become very important coping strat egies in the everyday lives of many working women.
Studies have shown that while advances in household technology have made life more pleasant and less arduous for all family members, and especially for women, these improvements have merely modified the type of house-work performed. Furthermore it is argued that improvements in household technology have been accompanied by a rise in standards of cleanliness and neatness, which have signifi cantly undercut time-saving gains from household technolo gies by increasing the total amount of time spent on house
26 Many of these technical changes depended on electricity and the introduction of indoor plumbing. In 1900 most U.S. homes had neither, but by the 1930's about 60% had electricity (enabling the use of electric refrigerators, stoves, and washing machines), and by the beginning of the Second World War 70% of homes had indoor plumbing (McDowell, 1983; Vanek, 1974, 1980; Whicker and Kronen- feld, 1986). 58 work (Ehrenreich and English, 1975; Oakley, 1974; Strasser,
1980; Vanek, 1980; Whicker and Jacobs-Kronenfeld, 1986).
Feminists argue that changes in household technology do not account for the increase in women's labor force participa tion, because the amount of time women spend on house work has remained about the same since the 1920's. Indeed time- budgets indicate that the amount of time spent on domestic tasks has remained remarkably stable over the last sixty years.
However what has changed is the distribution of tasks.
Time spent on cleaning and cooking has decreased, while time spent on child-care has increased (Vanek, 1974;
Sharpe, 1984). Furthermore studies indicate that women in the paid work-force do spend less time on domestic work than women not engaged in paid employment.27 Perhaps then these technological advances should be viewed in terms of enabling women to use their time differently, allowing more flexibility and making paid employment more viable. For instance, because standards have risen the temporal aspect of house-work has remained fairly stable; thus by lowering one's own standards one can decrease the burden of house work. Furthermore, tasks that previously had to be carried out on the same day, like laundry, can now be distributed
27 Studies estimate that women in paid employment do 16 to 28 hours of domestic work, compared with 32 to 47 hours for full-time homemakers (Meissner, Humphreys, Meis, and Schein, 1975; Pleck, 1983; Robinson, 1977; Walker and Woods, 1976). 59 over the entire week (England and Farkas, 1986)
2.3.5 Legislation and Popular Attitudes Affecting Women's
Paid Work
In the 1970's there was some decline in the extent of the gender division of labor in the labor force. This was largely due to women moving into "male" occupations, par ticularly the professions, and notably law, pharmacy, and medicine.
Most authors attribute at least some of this segregation break-down to legislative and legal changes such as the
1963 Equal Pay Act, the equal employment opportunities pro visions (Title VII) of the 1964 Civil Rights Act, and the creation of the Equal Employment Opportunities Commission to conciliate cases (although it was not until 1972 that it was given the power to take cases to court). In addition
Presidents Johnston and Nixon issued Executive Orders requiring government contractors to engage in Affirmative
Action, enforced by the Department of Labor's Office of
Federal Contracts Compliance Programs (OFCCP). Although attention initially focussed on racial discrimination, after much lobbying by women's groups the Equal Employment
Opportunities Committee (EEOC) began to consider sex dis crimination cases in the 1970's. Indeed the role of Femi nism played an important part in the introduction of the
legislation just described. For instance, in 1962, Presi 60 dent Kennedy. at the request of professional associations of women, lent his support to the principle of equal pay for equal work (Van Horn, 1988).
Table 9 shows the results of a study to examine the impact of these laws and regulations on firm policies since the 1970's (Survey of Firms by Bureau of National Affairs, cited by Freeman, 1980:337). It suggests that these laws and regulations may have forced many employers to permit women to enter into formerly male perserves (England and
Farkas, 1986).
There has also been a general ideological loosening of ways of thinking about paid work since the Second World
War. Men are beginning to realize that they may not neces sarily be in full-time employment earning a wage to support their family until retirement. Similarly many women no longer always expect to marry for life and never do paid work again (Beechey, 1985). Thus attitudes towards women working have changed drastically. Attitude and opinion surveys indicate that over time there has been a general trend towards the acceptance of working women. For
instance, in the mid-1930's only 15% approved of married women working, while 48% disapproved, and 37% said it depended on individual circumstances; by the late 1970’s
72% approved and only 26% disapproved (Erskine, 1980;
Oppenheimer, 1970).28
28 Interestingly, when Erskine looked at the gender differ- 61
Table 9
Changes in Personnel Practices Resulting from EEO Laws
Company programs and regulations % of companies Have formal EEO programmes 86 including Affirmative Action plan 96 (of those subject to OFCCP regulations)
Have had investigation or other action under Title VII 63
Have made changes in selection procedures for EEOC reasons 60 Testing procedures 39 Revised job qualifications 31 Application forms 20 Recruiting techniques 19
Have instituted special recruiting programmes For all minority workers 69 For minorities in professional/managerial 58
Have instituted programs to ensure EEO policies are implemented Communications on EEO policy 95 Follow-up personnel or EEO office 85 Training sessions on EEO 67 Periodic publications of EEO results 48 EEO achievements including performance appraisals 33
Have instituted special training programmes For entry-level jobs 16 For upgrading 24 For management positions 16
Source: Freeman (1980) cited in England and Farkas (1986:17)
Just as attitudes with regards to women working have
changed, so have attitudes about working mothers. The pre
vailing attitude used to be that children suffered if their
mothers worked, but these attitudes have changed as the
ential in positive attitudes to married women working, there was not a tremendous difference, although, in every instance, women were more likely to approve. 62 proportion of mothers working for pay has increased. For example Mason, Czajka, and Arber (1976) found that in 1964
54% of women surveyed agreed that a working mother could still maintain a close relationship with her children.
Within six years the percent of women holding this view had risen to 73%. People now say that they think that child care is a joint responsibility for both parents (although this is not necessarily played out in reality as mothers still tend to end up with the ultimate responsiblity). For instance a 1980 Harris Poll reported that only 19% of men and women think "raising children should be the responsi bility of the mother, not the father, whether or not she works" (Harris and Associates, quoted in Bianchi and Spain,
1986). So even though behavior has yet to catch up with professed beliefs, joint responsibility seems to be the ideal among most Americans.
Clearly dissatisfaction with the role of full-time home maker and the revival of feminism in the 1960's provided a new set of motivations for women to take on paid labor.
This would have been particularly true for older women with fewer domestic obligations and younger women who had not yet committed to full-time domestic roles (Van Horn, 1988).
Thus women in paid employment tend to no longer view their
labor force participation as intermittent or even as a com mitment secondary to their domestic role. Indeed women are 63 becoming more likely to search for full-time rather than part-time employment opportunities (England and Farkas,
1986). Historically, when women worked for pay -- prior to marriage, during economic depressions or wars, for instance
-- it was only on a temporary basis, whereas today it is more likely to be permanent.
Paid work provides women (and men) with opportunities to form and maintain social connections, giving their lives direction. And for some women, especially the married women with children, it provides an identity separate from their domestic roles (Glenn and Feldberg, 1984; Feldberg and Glenn, 1982). At the same time, societal changes such as the increased emphasis on equal opportunities and educa tional achievement have raised women's aspirations and expectations, especially among white, middle-class women.
Women with higher levels of education are more likely to be in the paid labor force and to stay there even if they become mothers. Although feminism has had the strongest impacts on women of the middle-class, it has also provided an ideology to justify the movement of women into tradi tional "male" occupations, to push for higher wages, better working conditions (such as the elimination of the more flagrant examples of sexist behavior), and to expand oppor tunities for upward mobility within traditional "female" occupations (Gerson, 1983; Van Horn, 1988). 64
2.4 THE FRENCH REGULATION SCHOOL
Many of the changing powers, wants and needs of>women outlined in the previous section are either of post-Second
World War vintage, or have become more ubiquitous during this period. Given that women largely have these changing powers, wants and needs by virtue of social structural change, care should also be taken to conceptualize this structural change. One way of doing this would be to draw upon the French Regulation School literature. Hence this section highlights the basic suppositions of this school of thought.
The French Regulation School approach (associated with the work of Aglietta (1979); Boyer (1979); De Vroey (1984); and Lipietz (1986)) conceptualizes the post-Second World
War era in terms of "Fordism" and "post-Fordism".2 9 The main endeavor of the Regulation School has been to analyze
"the articulation between mass production and mass consump tion which is the pivot of structural coherence within the post-war 'intensive' or 'Fordist' regime of production."
(Davis, 1984: footnote 2). Under the archetype of the
Fordist accumulation regime production takes the form of mass or batch production (particularly of consumer durable goods) giving rise to rapidly growing productivity gains; while on the consumption side there is an equally rapid
29 "Fordism" here does not refer to the common usage of a comprehensive application of scientific management tech niques to the labor force. 65 acceptance of mass produced consumer durables (Petit,
1986). Indeed accumulation is conditional upon the expan sion of workers private consumption of consumer-durable goods and vice versa. Set in place after the Second World
War, U.S Fordism has since been replaced by the very dif ferent social, political and economic configuration of
"post-Fordism" in the early- to mid-1970's (Davis, 1984).
The technical conditions for Fordism were the applica tion of "scientific management" and assembly line tech niques to the labor process in the production of mass- produced commodities. And the social conditions were the higher rates of labor productivity which enabled worker's real wages to rise without negatively affecting capital's profitability. As Perrons (1986) puts it:
In the postwar period many Western capitalist economies experienced unprecedented and sustained levels of growth. Underlying this wave of growth were new sources of energy and new methods of production. These methods, which included the introduction of "scientific management" or "Tay lorism" and the flow-line principle of Henry Ford, were particularly important in the newly developing consumer goods industries, and they raised labor productivity so that real wages and profits could rise simultaneously. These devel opments enabled a link to be established between the sphere of production and the sphere of con sumption so that some of the obstacles to, and instabilities within, the accumulation process were overcome, at least for a while (1986:248-249).
Thus a new economic stability emerged based on the link
forged between production and consumption/reproduction. In 66 order for capitalists to grant workers higher wages so that the mass private consumption of consumer durables could increase without decreasing profits, workers1 productivity had to increase. This was achieved through the shift from the formal to the real subsumption of labor to capital
(Marx, 1977:1019-1038). Under the formal subsumption, although the means of production already belonged to capi talists rather than the individual workers, the production process essentially remained in the same form as it had under pre-capitalism. Accordingly there was a unity of mental and manual labor, which is, for instance, symbolised in the notion of a craft-worker. However, such a unity represented a threat to capitals power, since it meant that the workers were more independent than they would have otherwise been. Moreover, reliance on unchanging labor processes made capital vunerable to bottlenecks in the labor force (bottlenecks both of a demographic nature and those emanating from legislation such as that limiting the length of the working day).
These pressures gave way to what Marx described as the real subsumption of labor to capital. This involved the purging of capitals reliance upon pre-capitalist work pro cesses in favor of their transformation by capital itself.
This revolution would, and did, emphasize the embodiment of workers1 knowledge in machinery and in the reorganization 67 of the technical division of labor and created a new form of worker control. In short, the workers were separated from their knowledge of the labor process, and that knowl edge became the property of a growing stratum of mental workers, particularly managers and engineers.
In order for workers to expand their mass private con sumption of consumer durables their bargaining power rela tive to capitalists needed to be strengthened. This was attained through increased State intervention via class and political struggle in the arenas of production and of the reproduction of labor power (Bonefield, 1987). So for instance provision was made for maintaining the worker's means of subsistence. Thus U.S. Fordism was founded on a limited social contract (or accord) between capital and labor, and legitimised through a minimal welfare state which made provisions for social security, minimum wages, unemployment benefits and so on (Bowles and Gintis, 1982).
Workers enjoyed increased levels of social security, which enabled them to maintain the same income (and, it follows, some expenditure) if they became unemployed (Aglietta,
1979; Perrons, 1986). Consequently workers became more financially secure and more able to purchase consumer dura bles and housing either directly or though the simultane ously expanding financial and long-term consumer credit institutions.3 0
30 Furthermore the increased profits of capitalists meant 68
Mediation of these consumption patterns by the State was fundamental. For instance, laws and tax relief for mort gages were introduced and the Federal Housing Authority was established in the 1930's and 1940's, making it easier to purchase housing.31 Indeed State intervention into educa tion was crucial given the technical and managerial exper tise required by the modified labor processes which provid ed the basis for productivity increases under Fordism.
Clearly Fordism is related to changes in gender ideolo gy, in the meaning of paid employment for women, and in their ability or desire to enter wage work. This is par ticularly true of the increased demand for female labor as a result of the expansion of "female-typed" employment such as clerical work.
that they were willing to invest and financial institu tions were willing to loan them money.
31 Davis makes a distinction between legislation which pro vided a regulatory framework for the emergence of an intensive regime of accumulation, and legislative pro grammes which underwrote specific structures of mass consumption and work-force mobility. Much of the clas sic New Deal legislation such as the various reforms of the national banking system, represent the former cat egory. The latter type include the Federal Highway Aid Act of 1956 and the National Defence Education Act of 1958. Such legislation determined the concrete forms of State intervention, as in suburbanization and higher education. "As such," Davis (1984:11) says, "they rep resent the systematic triumph of an extreme model of privatised mass consumption over the vaguely social- democratic concepts of public housing, mass transit and national planning espoused by war-time New Dealers." 69
2.4.1 Capital and Clerical work
Throughout the twentieth century, and particularly in the Fordist (and even the post-Fordist) era, clerical work has become increasingly more central to the economy. One indicator of this is the massive increase in clerical work ers both absolutely and as a proportion of the total labor force.
Table 10
>f the Clerical Labor Force, 1870-1980.
TOTAL CLERICAL AS A CLERICAL PERCENT OF (in 1000's) EMPLOYED PERSONS
1870 91 0.7% 1880 186 1.1 1890 490 2.1 1900 770 2.6 1910 1885 5.1 1920 3311 8.0 1930 4274 9.0 1940 4847 9.1 1950 7635 12.8 1960 9783 14.7 1970 13714 17.4 1980 18105 18.6
Source: Glenn and Feldberg (1984).
Table 10 indicates that between 1870 and 1980 the clerical worker portion of the labor force rose linearly, especially after 1940. Over the seventy year period between 1870 and
1940 the percent of all persons in paid employment who were in clerical work rose by 8.4 percentage points, whereas 70 over the 1940-1980 period the increase was 9.5 percentage points.32 This resulted largely from the rapid post-Second
World War increase in the functions and activities of the
State, and the shift in the economy to large, bank-financed firms and corporations with national and international mar kets. This affected clerical work in a profound manner.
First, functions such as record-keeping which had previous ly been incidental to clerical work, became it's very essence. Second, as offices became more complex, the demand for clerical workers increased both absolutely and disproportionally to growth in production (Glenn and Feld berg, 1984). Third, rather than being attached to a goods producing firm, as most offices previously had been, offic es were separated from the process of production, generally executing their activities through clerical labor (Braver man, 1974).
The largest users of clerical workers throughout the twentieth century in general, and after the Second World
War in particular, have been the Finance, Insurance, and
Real Estate and Government sectors. As Braverman (1974) notes:
32 However the rate of increase has slowed during the 1970's, although it should be noted that clerical work has continued to grow at a faster rate than the national employment growth rate and that clerical jobs are expected to grow by more than two million by 1990 (Davis, 1984). 71
Banks and credit agencies conduct only one mode of labor, the clerical, and below the managerial level the labor employed consists almost entirely of clerks who work in the offices and service workers who clean the offices... To a lesser degree, the same heavily clerical character of the labor process is true of law offices, and the offices of other institutionalized professions, advertising agencies, the publishers of books and periodicals insofar as they do not themselves do the work of manufacture, philanthropic and relig ious organizations, correspondence schools, agen cies for travel, employment, etc, and government offices for public administration.
As noted earlier in this chapter the single largest occupational category for women today is clerical work
(35.2%), and most clerical workers (80.1%) are women.
Indeed specific clerical occupations such as secretary and typist are among the most gender-segregated of all occupa tions. But this has not always been the case. Until about
1910 clerical work was done almost exclusively by men. So clerical work became sex-typed as "women's work" over the course of the twentieth century.
In nineteenth century offices male clerical workers worked in small, paternalistic, family-run businesses (Dav ies, 1975, 1982; Davy, 1986). Often working as personal secretaries, clerks handled all phases of an assignment, often doing tasks which would be considered managerial today, and indeed for a few men clerical work was a step ping stone to managerial positions (although these were usually the sons, nephews or grandsons of the owner; working-class men usually worked as clerks for their entire 72 work-lives). The paternalistic benevolence of the employer was the basis of worker control. Mutual loyalty and obliga tion provided the incentive for clerks to work harder, and some remained with the same firm for the duration of their work-lives (Dick and Morgan, 1987: Lockwood, 1958).
Table 11
The Feminization of Clerical Work, 1870-1980
FEMALE . FEMALE CLERICAL CLERICAL AS A PERCENT (in 1000's) OF ALL CLERICAL
1870 2 2.4% 1880 8 4.3 1890 83 16.9 1900 204 26.5 1910 677 35.9 1920 1601 48.4 1930 2223 52.0 1940 2549 52.6 1950 4597 60.2 1960 6629 67.8 1970 10233 74.6 1980 14502 80.1
Source: Glenn and Feldberg (1984).
Although employers were initially reluctant to hire women, because of their "distracting influence" and the threat that they posed to "male breadwinners",33 women came to be seen as an untapped source of educated and cheap labor (hence the rapid feminization of clerical work which
33 Lewis (1985) argues that the fear of female competition reinforced, and was reinforced by, the prevailing gender ideology that "a women's place is in the home", indicat ed that an analysis of the production and dynamics of gender ideology is crucial in the study of labor unions and employment in general. 73
Table 11 portrays).34 At the turn of the twentieth century women were more likely to have graduated from high school than men were. And while men with high school diplomas had a wide range of job options open to them, this was not true for women (Oppenheimer, 1970; Kennedy, 1979; Kessler-
Harris, 1982). Excluded from most "professions" and col lege level teaching (Davies, 1975, 1982), most working women were employed as teachers or governesses if they were middle-class, or as domestic servants or factory workers if they were working-class (Kessler-Harris, 1981, 1982).
Clerical work offered women an attractive, viable, alternative. It was mental rather then manual employment which took place in a safe, clean work environment, along with shorter working hours and direct personal contact among workers and between workers and managers. Clerical workers also had higher status, greater job security and
34 Firms also introduced machines such as the typewriter. Patented in 1873 by Remington, numerous authors contend that the typewriter facilitated the entrance of women into clerical work (Davies, 1975, 1982; Davy, 1986; Goldberg, 1983). As a new occupation, typing was ini tially "gender-neutral", so women were not accused of stealing "male bread-winners" jobs, and second, Reming ton decided to train women to demonstrate the new machines. Thus typing soon became labelled as "women's work". This is illustrated by the increase in, and fem inization of, typists and stenographers. In 1870 only 4.0% of the 154 typists and stenographers were women, by 1910, 76.7% of the 112,600 typists and stenographers were women. Accordingly by 1935 FORTUNE magazine declared that "woman's place was at the typewriter" (Davies, 1975, 1982). Other office "machines" intro duced have included counting machines in the 1950's and, more recently, computers and visual display terminals (VDT's). 74 better opportunities for promotion than other jobs open to women at that time, and they enjoyed a fixed, weekly rather than hourly wage (Glenn and Feldberg, 1984). Indeed cleri cal work paid more than most other "women's work": typists and stenographers were paid $6-$15 per week, whereas domes tic servants received $2-$15, factory operators $1.50-$8
(Davies, 1975, 1982), and there is some limited evidence that clerical work paid more than teachers (Rotella,
1981).35
Until the Second World War, most of the female clerical workers were single, living with their parents and only intended to work until they got married.36 Indeed recruit ment and office functions were actually based on this assumption, and it was widely used as justification for
35 Many employers routinely used family background as a criterion for hiring a woman. Most of the first women clerical workers were white, educated, and from American born, middle-class backgrounds hired for their "breeding and respectability" (Kennedy, 1979). Indeed the high school education requirement for women clerical workers was intended to exclude working-class women, Blacks and immigrants from the office. Generally working-class women became a larger proportion of the clerical work force in the 1920's, but Black women found that they were excluded until the 1960's and the advent of the Civil Rights Movement (Kessler-Harris, 1981, 1982).
36 It should be noted that generally women only wanted to work until they married. Marriage and domestic life were seen as very desirable, and marked the start of their proper and most fulfilling station in life. Marriage rescued them from the tedium of clerical work and the controlled environment of their parent's home. Indeed marriage was regarded as a form of independence because they could have some autonomy and control in their own homes (Sanderson, 1986). 75 ghettoising women into low level positions, not promoting them, and paying them about half the salary of equivalent male clerks, because as unmarried women their fathers, sup posedly, were subsidizing them and they would resign once they married (Davy, 1986). Furthermore, until the Second
World War many firms had a formal marriage bar for women requiring them to resign if they got married (Davy, 1986;
Sanderson, 1986).37 Sanderson (1986) documents an inter esting method used by the British Civil Service to deter women clerks from remaining employed after marriage. Many of the women she interviewed mentioned receiving "dowries” when they resigned:
The "dowry" which (the women) mentioned, remem bering the exact amount, was actually a gratuity in lieu of pension. It was paid to the women when they resigned to get married, thus making them a rather good "catch" for any aspiring man. The gratuity was paid at the rate of one month's pay for each year of service, but not exceeding twelve months pay. It was not financially worth while for a woman to stay in the (Civil) Service beyond twelve years, as the gratuity would cease to accumulate after this. After twelve years a women would only get a proper return on her pen sion contributions if she stayed in the Service to retirement, and stayed single (Sanderson, 1986:115).
37 Besides the clear gender ideology, especially during the Depression, that working women were putting "male bread winners" out of work, there was an equivalent ideology that working women who were married were taking work away from more "worthy" single women. Additionally mar ried women in paid employment faced the social stigma that assumed she either had a husband who was unable to "keep" her or that she had suffered some personal mis fortune. So if working women were married, they had an easier time if they kept the fact from their employers and co-workers (Davy, 1986; Hartman Strom, 1987). 76
The advent of women clerical workers marked a transition in the organization of offices. Unlike most men that they replaced they did not work as personal secretaries, and very few found themselves being promoted into managerial positions. Instead many found themselves constituting a
"secretarial pool" which any (male) executive could draw upon, doing tasks which had been sub-divided into a series of steps, each completed by different women to produce
"maximum efficiency with minimal training" (Kessler-Harris,
1982).38
More recent developments in the computer industry have enhanced this form of worker control. The invention of the silicon chip and other related innovations has facilitated the development of small, inexpensive data-processing sys tems, putting automation within the budgets of even the smallest of firms (Fox and Hesse-Biber, 1984; McNally,
1979). Some observers argue that the distinction between clerical work and blue-collar work has become increasingly blurred as clerical work is being deskilled39 by automation
38 As early as the 1920's and 1930's attempts were made to "rationalize" office functions through the borrowed "scientific management" techniques of Taylorism (exten sive time and motion studies were conducted down to the minutest detail).
39 Deskilling involves a fine division of labor with each subdivision standardized into a series of fragmented, repetitive steps, each assigned to different workers. A pace of work is encouraged which forces the workers to work as fast as possible without loss of efficiency, under close depersonalized supervision to prevent slow downs, sabotage, and error. Accordingly, this process 77 and standardization principles traditionally applied to manual work (Braverman, 1974; Feldberg and Glenn, 1988;
Glenn and Feldberg, 1982; Shepard, 1971). Indeed Glenn and
Feldberg (1982) argue.that clerical work has become "prole- tarianized" due to deskilling.40 As a federal task force reported:
Secretaries, clerks and bureaucrats were once grateful for having been spared the dehumaniza tion of the factory ... they had higher status than blue-collar (workers). But today the clerk ... is the typical American worker ... and such positions offer little in the way of prestige ... imparting to the clerical worker the same imper sonality that blue-collar workers experience in the factory (Work in America, 1973, quoted in Glenn and Feldberg, 1982:202).
By restructuring the work-process in this manner the manager's control over the work-process is enhanced, reli ability and standardization are assured, and the variety and level of skills required for clerical workers are reduced. On the other hand, automation enables workers to be more productive and improve the quality of their work.
alienates the conception of a task from it's execution, serving to prevent the immediate producer from gaining complete knowledge of the entire production process. So for instance, Garson (1985) found that some clerical workers now face impersonal controls such as clocking- in, their supervisors checking their number of key strokes, mistakes, and their time away from the key board (Garson, 1985).
40 Following C. Wright Mills, Glenn and Feldberg (1982) define "proletarianization" as a "shift in middle-class occupations toward wage workers, in terms of: income, property, skill, prestige or power, irrespective of whether or not the people involved are aware of these changes" (Mills, 1956, quoted in Glenn and Feldberg, 1982: 203). 78
For some clerical workers the important attraction of acquiring and honing their computer skills are, first, their belief that such skills will facilitate their promo tion prospects, and second, that their new computer skills will be reflected in their salaries. However, indications point to a limited realization of promotion prospects for women clerical workers, and that many employers (and indeed, women clerical workers) believe that on-the-job training is a sufficient substitute for additional pay
(Murphree, 1987).
Due to the prevalence of occupational segregation by gender within offices, numerous authors fear that the work er alienation and redundancy associated with the automation of factory work, will have a disproportionally negative impact on women clerical workers, but may benefit men because they will dominate the newly created, highly skilled, technical and professional jobs, and because auto mation may further concentrate control in the hands of
(male) senior executives and management.
Basically the effect of automation on clerical work is still very much a subject of debate. For instance, some researchers have found job loss to be concentrated among low skilled clerical workers, implying an upgrading of positions (Roessner, Mason, Porter, Rossini, Schwartz, and
Nelms, 1985; Shepherd, 1971); while others believe that 79 automation results in the elimination of skilled jobs, leading to the degrading of clerical work (Feldberg and
Glenn, 1988; Glenn and Feldberg, 1982; West, 1982). Simi larly, while some writers have found less task fragmenta tion as technology becomes more sophisticated (Adler, 1983;
Appelbaum, 1987; Matteis, 1979; Sirbu, 1982); others sug gest that job content is narrowing and that worker autonomy is being eroded (Feldberg and Glenn, 1988; Glenn and Feld berg, 1982; Greenbaum, 1979; Murphree, 1982, 1987). Final ly, although in all cases women experienced the largest job loss, some researchers found that after automation women were downgraded to lower skilled activities (Glenn and
Feldberg, 1982; Murphree, 1987), whereas others have noted that woman clerks seemed to benefit from the new labor pro cess (Matteis, 1979).
One further form of worker control that observers have brought to light is one based simply on gender, exploiting prevailing stereotypes that women are more oriented to pleasing others, more honest, and less mercenary than men
(Glenn and Feldberg, 1984; Lamphere, 1985; Littler, 1982).
As Glenn and Feldberg remarks
Clerical workers are sometimes thought to be treated "better" than production workers. On closer inspection this "better" treatment may disguise closer, more restrictive controls. Lan- ger (1972) shows how the New York Telephone Com pany "rewards" workers while maintaining control and low wages. The women are "treated" to candy on holidays, jewelry on their anniversaries with the company, and appliances for recruiting new 80
employees. "Niceness" is stressed throughout the company to create a pleasant atmosphere. This "niceness", which extends to sharing workloads, helps the women cope with the strains of the con stant supervision and rigid formats; it thus enables the company to continue imposing "unrea sonable" demands ... Orders to women are cloaked in the guise of personal requests, which they find difficult to refuse ... Even when the man agement is aware that its employees see through its rationale, they rely on them to act like ladies and to continue to be loyal, dependable, and polite (Glenn and Feldberg, 1982: 212).
2.5 THE COLLAPSE OF FORDISM
Explanation for the collapse of U.S. Fordism in the early- to mid-1970's lies in the exhaustion of the possi bilities for increasing labor productivity and profitabili ty under Fordist labor processes (Perrons, 1986). There was a relative saturation of the consumer-durables market and growing competition, especially from West Germany and
Japan, and particularly in many of the traditional manufac turing industries such as steel, autos and rubber.
Attempts to increase productivity by increasing the inten sity of work met with the resistance of workers, whose bar gaining power vis-a-vis capital had been strengthened throughout the Fordist era. One manner in which some firms reacted, especially trans-national corporations, was to step-up the locating of their more routine units of produc tion "off-shore" in order to maintain the combination of literate, skilled labor and low wages. Another reaction 81 has been to expand the Fordist dynamic in the U.S. itself through more automated or "neo-Fordist" (and often "des killing11) labor processes. These include electronic infor mation systems with automatic feedback mechanisms which have been especially prevalent in the growing service sec tor .
Accordingly, this section will explore some of the major characteristics of post-Fordism as they relate to the paid employment of women, especially those employed or employa ble in clerical work. First the decline in real wages will be considered, followed by an examination of the relation ship between the "Middle-Class Dream" and the increased significance of paid labor for women. This section is fol lowed by a discussion of the recent increase in part-time and temporary paid employment.
2.5.1 Declining Real Wages and the "Middle-Class Dream"
During the Fordist era real (inflation adjusted) wages grew by approximately two to three percent per year, and for a growing proportion of the population the dreams of a middle-class life-style became a reality (Levy, 1987). An important characteristic associated with the collapse of
Fordism, however, has been the stagnation of real wages: indeed some argue that real wages have declined (Bluestone and Harrison, 1986; Thurow, 1987). Although there is not complete agreement among scholars, the evidence indicates 82
that the growth of real (average hourly) wages has slowed considerably since 1973 (Loveman and Tilly, 1988).41 Table
12 shows male and family income growth from 1949 to 1981.
In the first three time periods men's median income grew
steadily for each cohort; in the later periods they stag nated.42 In 1949 the average 30-year old man would have
earned $12,000; ten years later he would have earned
$18,958; and by age fifty he would have earned $24,421. In
1973 the average 30-year old man earned $23,579, but by
1984 he was only earning $23,418. Indeed a man who became
forty in 1973 earned 14% less by 1984, as opposed to 25%
more as his father would have done in the 1950's and 1960's
(Levy, 1987).
Some authors have also pointed to the growing inequality
in the income distribution of families. The traditional
income "pyramid" has been replaced by a new income "hour
glass". As Parker (1981, quoted in Davis, 1984) points
out:
41 Loveman and Tilly (1988) have reviewed the research designs and results of twenty-six key papers in the declining real wages (and job quality) debate, includ ing: Bluestone and Harrison (1986); Freeman (1986); Rosters and Ross (1987); Lawrence (1984); Rosenthal (1985); Stanbeck and Noyelle (1982); and Thurow (1987).
42 Women's median income growth shows a similar pattern: annual real income (for full-time employment) rose by 9.2% between 1955 and 1959, 54% in the 1960's, and dropped by 1% between 1973 and 1984 (this figures will, of course, have been affected by the rapid increase of women going out to work). 83
Table 12
Male and Family Median Income Growth, 1949 to 1984
(in 1984 dollars)
Men's Median Incomes age 1949 1959 1969 1973 1984 25-34 years $12,000 16,916 22,593 23,579 17,538 35-44 years $12,858 18,958 25,628 28,118 23,418 45-54 years $11,987 17,290 24,421 27,279 24,132
Median Family Income: $13,540 19,300 26,700 28,200 26,433
Source: Levy (1987:47, 87)
Middle-class families — truly in the middle -- are disappearing, displaced by two-income, two- person households ... on the top of the hour glass, and single-earner blue-collar families, clerical singles, women in many jobs, the welfare poor, and the retired on the bottom".
For instance, Thurow (1987) notes that in 1985 the top 20% of all families received 43.5% of the total income in the
United States, the highest level ever recorded since data were first collected in 1947. Conversely, the income share of the lowest 60% of families fell to a record low of
32.4%. Bradbury (1986) found that between 1973 and 1984 the percent of all families with incomes above $50,000 (in
1984 dollars) rose slightly from 14.9% to 15.6%, whereas the percent of all families with incomes of less than
$20,000 rose from 32.1% to 36.4%. At the same time the percent of "middle-class" families (those with a median family incomes of $20-50,000) fell from 53% to 48%.
While today's middle-class dream does not carry a precise price tag, it exists in popular con- 84
sciousness and has come to include a single family home, one or two cars (including one new car), a washing machine and dryer, a dishwasher, a color TV, raising and educating children, pro viding for a period of retirement, and so on (Levy, 1987:6)
Accordingly since the mid-1970's it has become increasingly difficult to obtain and maintain the "Middle-Class
Dream".43 For instance, owning a home, a staple for the middle-class standard of living, has become less economi cally feasible for many people. In the "traditional fami ly" of the 1950's, for only 14% of his earnings the typical
30-year old working man could make monthly payments on a median priced house; in 1973 this figure rose to 21%, and by 1984 it stood at 44% (Levy, 1987).
The increased labor force participation of women has become a major way in which middle-class American families have attempted to maintain their standard of living and level of consumption (Currie, Dunn and Fogarty, 1980).44
4 3 Those who had already attained the dream had some pro tection (such as job seniority and fixed mortgage pay ments) and people who had already retired found that their Social Security benefits were regularly adjusted for inflation (Levy, 1987).
44 This does not explain why labor force participation of wives increased during the 1960's, when real individual wages and median family incomes were still increasing (which, theoretically, should have been a strong work dis-incentive for wives). Besides proposing that it may have been related to the re-emergence of feminism, Van Horn (1988) suggests that a more detailed examination of men's wages provides a partial explanation. Growth in salaries for white-collar workers (except those in the most senior positions) failed to match the relatively greater increases in wages for blue-collar workers. Thus white-collar families experienced a relatively low 85
Certainly as Table 13 indicates the proportion of wives in paid employment has increased steadily since 1950. Indeed over the same period the paid labor force participation of husbands has declined, partly due to the recent trend towards early retirement and partly due to the decline in well-paid blue-collar men's jobs associated with the col lapse of Fordism.
Table 13
Spouse in the Paid Labor Force, 1950-1980
YEAR Wives Husbands 1950 23.8% 91.6% 1960 30.5 90. 7 1970 40.8 86.9 1980 50.2 81.0
Source: Sternlieb, Hughes and Hughes (1982:49).
Further evidence is provided by Table 14 which reports the economic effects of a wife in paid labor on median family income. The first two columns are a comparison of the median family income of families with one (husband) wage earner to those with two (wife and husband) wage earners.
Essentially the median family income of families with two wage earners has continued to grow, albeit slightly, in the post-Fordist era, while the converse is true of families where the wife is not in paid labor. The third column
er level of improvement in family income in comparison with blue-collar families. Van Horn suggests that this may have induced more middle-class women to enter paid labor in order to contribute to the greater relative improvement of their family's income. 86
Table 14
Median Family Income: Presence of Wife in Labor Force
(in 1980 dollars)
Average Percent Added Income of Wife in Paid Wife Not in Paid Families with YEAR Labor Force Labor Force Working Wife 1950 $13,700 $11,346 — 1960 19,197 15,358 25% 1970 26,330 19,742 32% 1980 26,879 18,972 42% 1983 —-- 47%
Source: Sternlieb, Hughes and Hughes (1982) and Van Horn (1988:174) illustrates that the average percent of added income that working wives provide their families has grown tremendously since 1960. This strongly suggests that two-wage earner families have become increasingly necessary in order to achieve continued real improvement in family income. More over it is quite feasible that the slower decline of real median family income compared with real individual income
(see Table 12) is due to the increased labor force partici pation of wives (Klein, 1984; Sternlieb, Hughes and Hughes,
1982; Power, 1988).45
45 In part the income increases documented here are due to higher wages earned and more hours worked by women since 1950. But they are also associated with social changes, such as more wives of higher income husbands have been going out to work. Among wives who are in paid employ ment, well-paid wives tend to be married to well-paid husbands, which would imply an increase in family income inequality. But because the wives of well-paid men are less likely to work, and because the range of women's earnings is not as great as the range of men's earnings, wives' earnings tend to deflate inequality. Consider 87
2.5.2 "Flexibility": Temporary and Part-Time Work
The hallmark of recent labor policy has been the pursuit of what is -- euphemistically -- described as "flexibili ty". This seeks to enhance "competitiveness" by giving greater power to management through modifying the condi tions and terms of employment to what management consider to be the requirements of the production process and the conditions of the market (Bluestone, Tilly and Harrison,
1986; Piore, 1986). So given falling profits and increased
economic uncertainty, firms have been paying more attention
to cost containment. This helps to explain the attraction
of, and recent growth in, part-time and temporary workers.
Instead of paying full-time workers over-time, temporary
and part-time workers can be used. Indeed these workers are
seen as "economic buffers" to be taken up during business
upswings and laid off during downswings.46
two examples. In 1984, 58% of women married to men earning more than $35,000 were in paid employment earn ing an average of $12,600 a year. So the average women in this group (including homemakers) earned $7,300 or 14% of her husband's earnings. In the same year 68% of women married to men earning $7,000 to $15,000 worked earning an average of $8,500 a year. Thus the average woman in this group earned $5,100 or about 40% of her husband's income. Thus married women's earnings have relatively greater effects on lower-income families, so reducing family income inequality (Levy, 1987).
46 Less quantifiable advantages are accounts which intimate that two part-time workers tend to be more productive than one full-time worker. Because part-time workers will be less susceptible to fatigue and stress, they reportedly have lower rates of absenteeism, and even their higher rates of turnover (which some argue increase costs because it requires an increase of 88
Since 1968 part-time work has been growing at a faster rate than full-time work. In 1954 (the first year that comprehensive data were collected) 15.4% of all non- agricultural workers were employed either part-time or tem porarily. By 1984 this figure was 22%, 26.8 million people, or an increase of approximately 4% per year, double the rate for full-time workers.
At the same time the number of temporary-help agencies has increased dramatically. From employing only a few thousand in 1946 to five million by 1985, temporary-help agencies have become one of the fastest growing industries in the United States. According to the Bureau of Labor
Statistics, temporary help agencies averaged a 11% per annum growth between 1972 and 1985, compared to 2.1% p.a. for all non-agricultural jobs.47 Significantly the fastest growth rates in the economy as a whole are within the very industries with the largest proportions of part-time work ers -- low-level services, finance, insurance, real estate,
resources allocated to recruitment and training of new personnel) can be interpreted positively in that it could be acting as a form of worker control by inhibit ing demands for promotion and increased wages (Flint, 1977; Fowler 1977; and Samuelson, 1980, cited in Soko- loff, 1987).
47 NOTE: 1. This does not include those people hired on a temporary basis directly by the companies themselves. 2. This trend is expected to continue particularly since new regulations governing the employment of tempo rary federal workers went into effect in January 1985. These new regulations state that temporary jobs may be extended into higher skill categories and can last for up to four years. (Smith, 1983; 9-to-5, 1986). 89 wholesale and retail trade (Sokoloff, 1987). Part-time and temporary workers are particularly suitable to these indus tries as they are characterized by daily (for example, fast food restaurants) or yearly (for example, insurance indus try) business "peaks" and "troughs".
Most part-time workers are women,40 and in 1983 women occupied nine million (about 70%) of all part-time jobs.
Two-thirds of part-time workers are women employed in cler ical work, sales, and low-level services. Furthermore 62% of all temporary workers are women, most of whom do cleri cal work. Thus low-level, part-time jobs have flourished among the fast growing female dominated sectors of the economy to such an extent that in 1978 75% of all low-level service workers and 50% of clerical workers worked part- time .
Part-time and temporary work has had a considerable effect on clerical work: 24.7% of all clerical workers are employed on a part-time or temporary basis, making it the largest concentration of temporary employment (seventy per cent of all temporary workers are clerical workers) and the third largest concentration of part-time workers. The var iation in the types of part-time and temporary clerical work available can range from working regularly but for less than thirty-five hours a week, to being part of a
48 Although before the 1950’s men far outnumbered women in part-time jobs. So part-time work has become increas ingly feminized in the last fifty years. 90 firm's internal labor pool of experienced temporary employ ees. As Smith (1983) explains:
The variation in types of part-time or temporary employment are various ranging from working regu larly less than 35 hours per week, to being part of an internally regulated labor pool (available on-call in that company when needed), to well- known "temporary help" arrangements offered by such agencies as "Manpower," "Temporaries," "Kel ly Services" and so forth.
The temporary help services in many ways dem onstrate the organization and logic of part-time structures. The primary way in which temporary clerical workers are used is to employ them dur ing peak business activity, thus decreasing the need for regular employees to work overtime ...
Temporary workers notwithstanding, regular part-time employment rates of clerical workers have also increased due to rigorous tightening, or paring down of hours involved in specific jobs. For example, at J.C. Penney's, a major department store chain, part-time clerical work ers are used to cover peak periods of business activity. Many full-time positions have been cut in half to part-time positions in order to release more people for these periods. In addi tion, Penney's has a "pool of people" to call when needed; thus, their temporary workers are company employees who are familiar with docu ments, office procedures and management (1983:4-5).
Indeed some firms are even re-organizing their labor force to employ more part-time workers so they do not have to pay benefits, pensions or allocate money for training and promotion. There is even evidence that some firms have re-organized in order to use temporary workers, allowing for them in their annual budgets rather than using them when unforseen circumstances arise (Appelbaum, 1987). 91
Most part-time workers are located in low-skilled entry level positions. Such positions are usually associated with low pay (one estimate is that part-time workers earn
29% less an hour than full-time workers in the same job
(Sokoloff, 1987); another is that they earn $4.50 per hour compared to $7.80 an hour for full-timers (Appelbaum,
1987)); limited benefits (workers employed for less than thirty-five hours receive few, if any, benefits); limited training and poor job security.
Thus part-time work heightens the exploitation of women as cheap labor. In fact part-time work is perhaps the clearest expression of women as a "secondary labor market"
(Baron and Norris, 1978; Doeringer and Piore, 1971)49 and a
"reserve army of labor" (Beechey, 1977; Bruegal, 1979;
Milkman, 1976), earning "pin-money" or incomes "secondary" to their family's budget (Beechey, 1977).50 Part-time jobs
49 As opposed to the "primary labor market" with full-time employment of skilled workers who receive relatively high wages, good benefits and job security, who have some control over their work situation and who are con sidered to be "dependable".
50 Domestic obligations are the most common reason cited by women as to why they work part-time, although this rea son has begun to decline in importance since the 1960's (Deuterman and Brown, 1978, cited in Appelbaum, 1987). The propensity of women to be employed part-time is related to their marital status and the age of their children. For example, a higher proportion of employed married women than of employed divorced, separated, and widowed women work part-time. Among employed married women, those with children under age eighteen are most likely to work part-time, while the reverse is true of divorced, separated, and widowed working women (Barrett, 1979). However choice of part-time work cannot be 92 have been depicted as beneficial to women, offering them the flexibility to fulfil their domestic roles. According ly some firms have explicitly organized their part-time shifts around maternal and marital statuses. Some, there fore, offer "mother's jobs", arranged to coincide with school hours and holidays, while others have the "house wife's shift" where women can pick their schedule according to their domestic needs (Smith, 1983). Indeed in many analyses of women's part-time paid work the benefits accorded to capital are often overlooked in favor of laud ing the benefits of part-time work for women. Yet as
McNally (1979) has pointed out:
So long as agencies and temporary workers exist there is no pressure upon the employers to adopt more flexible attitudes themselves. An employer who makes a concession to one permanent worker is obliged to extend this facility to all the others - let one married secretary come in at 9:30 and they will all want to (1979:98).
The existence of a part-time and temporary labor force may reduce the pressure on employers to accommodate the special needs of women workers, such as on-site child-care and parental leave. It also means that employers do not have to worry about working conditions, salary, promotion, or even on-the-job training for their workers (Fox and Hesse-
assumed, in all cases, to reflect preference as, for instance, women of young children are clearly con strained by the lack of affordable, quality child-care. It is currently estimated that there is one child-care position for every ten children who require placement and that the average cost of which is about $4,000 per year for two children (Appelbaum, 1987). 93
Biber, 1984).
2.6 SUMMARY
This chapter has examined the literature exploring ques tions of social structure as they relate to the paid employment of women, particularly in clerical occupations.
The first section argued in favor of the socialist feminist view of the intertwined nature of patriarchy and capital ism. This was followed by a discussion of the various forms of the gender division of labor in the twentieth cen tury. It was pointed out that the relationship between patriarchy and capitalism is a dynamic one, as the massive increase in the paid employment of women since the Second
World War demonstrates. In an attempt to explain why this growth occurred, women's changing powers, wants and needs were outlined. These were conceptualized in terms of changes in the social structure by borrowing the concept of
"Fordism" from the French Regulation school. One way in which Fordism was seen to have had a clear impact on womens' paid employment was in terms of the rapid increase in the demand for clerical work. Furthermore, it was argued that the recent increase in the paid employment of women, particularly those with young children, is related to the collapse of Fordism. This is manifest, for
instance, in the now commonplace situation whereby women's 94 wages have become vital to the maintenance of their fami lies' standards of living. CHAPTER III
THE LITERATURE: QUESTIONS OF SPACE
3.1 INTRODUCTION
This chapter reviews the literature examining questions of space as they relate to the post-Second World War growth of women's paid employment in general and in clerical work in particular. The first section provides a conceptualiza tion of an important theme in this dissertation, that of spatial entrapment. This concept is at the heart of a good portion of the geographic research about women's paid employment conducted thus far. Essentially, when applied to women in paid employment, the concept of spatial entrap ment suggests that in their attempts to combine their vari ous roles, women are confined to a spatially limited set of options. Thus, although women may have been spatially entrapped in the past, it has only come to be perceived as a social problem in the more recent past due to the increased numbers of women negotiating multiple roles.
As the aim of this dissertation is to explore the links between changing spatial patterns of clerical work and the rapid increase in women's paid employment since the Second
- 95 - 96
World War, the work by Kristin Nelson (1986a, 1986b) on the
changing geography of the supply of female clerical workers
is of central importance. She provides one of the few
efforts to explore the link between the changing spatial
patterns of clerical work and women's paid employment; and,
additionally, in her explanation of these patterns, she
implicitly employs the concept of spatial entrapment of
women and its implications for office location.
3.2 SPATIAL ENTRAPMENT: A CONCEPTUALIZATION
Despite the rapid post-Second World War increase in
women entering the paid labor force, the traditional gender
roles of women as "natural" caretakers of the home and fam
ily has remained remarkably unaltered. Thus when women
combine- their roles as mother, domestic manager, partner
and wage earner they face time squeezes and need to recon
cile their various roles within a finite period of time.
In addition women negotiating multiple roles may find that
they face spatial constraints such as poor public transpor
tation and limited opportunities for affordable quality
child-care at convenient locations. (Coser, 1982; Fava,
1980; Markusen, 1981; Rothblatt, Garr and Sprague, 1979;
Wekerle, 1981).
However the multiple roles of women imply no necessary
relationship to space. This is so whether we consider all 97
the spatial relations such roles involve or for any of them
taken separately. Instead the relationship to space should be seen as one of time budgeting subject to a variety of
contingent conditions, such as the availability of child
care. Essentially time budgeting refers to detailed
accounts of the way in which people spend their time.
While this can provide a useful description of people's
actions, a spatial dimension is required in order to anal
yse people's activity patterns (Tivers, 1985). One
approach to time budgeting which does involve a spatial
dimension is that of "time geography". According to Hag-
erstrand (1967, 1970) every person follows a daily (or
monthly, or yearly) time-space path. Along this time-space
path are stations (which are fixed in time and space) where
the person stays for a period of time (for instance the
hours spent in paid employment each day) (see Figure 1).
Moving from one station to another is often hampered by a
number of constraints. Hagerstrand identifies three types
of constraint: "capability constraints", "coupling con
straints", and "authority constraints". The first has both
a biological need and an enabling component. For instance,
people need to eat and sleep, and having access to a car
enables a person to travel a greater distance from home
than people who are reliant on public transportation.
Coupling constraints refer to the temporal and spatial 98
Distance which can be travelled within allotted time: Time
by a pedestrian by car before returning home
work during lunch- \ . break
work before work
homle homle home W Distance
"JANE'S" DAILY PRISM Time
Distance
figure 1 99 restrictions faced when, for example, a person needs to coordinate her/his activities to fit with other people's schedules, and with the operating hours of services. Final ly authority constraints refer to the limitations and con trol of access due to customs, laws, rules and expecta tions: for example, work-places requiring union membership, and limited places in a child-care center.
Using this approach time and space constraints are consid ered in relation to the activities in which an individual wishes to partake. In particular time-space paths can be graphed out in terms of daily prisms showing "time-space walls" which indicate the distance which can be covered within an alloted time and which confine the individual's activities (Giddens, 1985; Gregory, 1985; Hagerstrand,
1967, 1970; Pred, 1985; Palm and Pred, 1978; Thrift and
Pred, 1981).
Feminist geographers have argued that gender ideology must be acknowledged as being a key constraint on both women and men because it pre-determines gender roles and so differentially ascribes particular tasks and activities by gender. Women's gender roles serve to restrain their behavior, limit their activities and confine them to a smaller geographic area than men (Cichocki, 1980; Miller,
1983; Monk and Rengert, 1982; Tivers, 1978, 1985; Palm and
Pred, 1978; Pickup, 1984). Thus time geography studies 100 could be employed in order to examine the spatial and tem poral constraints of women when negotiating multiple roles.
For example Palm and Pred (1978) provide the instance of
"Jane", an unmarried women with a two-year old child, con sidering two job offers. Her time-space prism is presented in Figure 1. "Jane" cannot choose work-place two (W2)
(although it offers a better salary and is a more challeng ing job) because the location and operating times of the child-care center are such that she would arrive late for work in the morning, and in the evening she would arrive at the child-care center too late to pick up her child (Palm and Pred, 1978).
By assuming multiple roles, women, especially the moth ers of young children and those living in less accessible neighborhoods, face more severe time-budgeting problems.
Essentially their gender role limits their activities to a spatially limited range of options. In consequence, it is argued, they form a "captive" labor supply, confined to a small job search area close to home, obliged to accept lower-paying jobs or part-time employment instead of the full-time work that they desire. Thus the remainder of this section examines some of the ways that historically received and socially constructed gender roles have inten sified the problems of time-space budgeting for women, par ticularly through their implications for the mobility char 101 acteristics of women and men, the inequality inherent in human built environments, and residential location deci sions .
3.3 IMPLICATIONS OF TIME-SPACE BUDGETING
This section elaborates on some of the concepts intro duced in the previous section. Specifically it considers how historically received gender roles intensify the prob lems of time-space budgeting through their implications for the mobility characteristics of the genders, the human built environment, and residential locations.
3.3.1 Mobility and Gender
Married couples will often allocate household resources along traditional patriarchal lines in such a way that the husband has more access to the family car leaving the woman to rely on slower, less flexible public transportation
(Dix, Carpenter, Clarke, Pollard, and Spencer, 1983; Pick up, 1984). As such, studies have found that when compared to men, women are less likely to own or have access to a car (Guiliano, 1979; Gurin, 1981; Hanson and Hanson, 1980,
1981), and they are less likely to hold a driver's license
(Fox, 1983; Hartgen, 1978; Pucher, Hendrickson, and McNeil,
1981) making them more dependent on public transportation
(Hanson and Johnston, 1985; Rutherford and Wekerle, 1988;
U.S. Department of Commerce, 1979; Fagnani, 1983). 102
In Baltimore it was found that 74.5% of the women and
83% of the men commuted to work by car, while 19.5% of the women and 13% of the men rode the bus to work (Hanson and
Johnston, 1985). Indeed not only are women less likely to travel to work by car than men, but when they do they are more likely to be a passenger or in a car-pool than men are
(only 20% of U.S. workers car-pool) (Ericksen, 1977; Pick up, 1984; U.S. Department of Commerce, 1979). For instance, Ericksen (1977) found that 39% of the men, but only 13% of the women in her sample drove themselves to work, whereas 12% of the women and only 6% of the men tra velled to work as a car passenger. Comparing modes of transport, Ericksen (1977) found that those women who drove their own cars to work had the shortest travel-times, whereas those women who car-pooled or used public transpor tation had the longest work-trip times.
3.3.2 The Human Built Environment and Gender
Feminist-inspired work in the environmental disciplines of geography, urban and regional planning and architecture notes that the particular form that gender relations and roles take at any point in history will be fossilized into the concrete form, design and location of residential are as, work-places, transportation networks and the overall layout of cities in general (Fava, 1980; Bowlby, Foord,
McDowell, 1986; Freeman, 1981; Hayden, 1981; Keller, 1981; 103
Kessler-Harris, 1981, 1982; Mackenzie, 1983, 1987; Macken zie and Rose, 1983; Westwood, 1984). As Bowlby and McDo well point outs
At any particular time in history, ideas about gender and the family help to determine the phys ical design and location of dwellings, of places of work, open spaces and other designed settings. These environments then support and reinforce the ideas of gender and family that generated them. The design of environments translates into physi cal form society's expectations of what activi ties should take place where, who should pursue those activities, and how they relate to one another. Sometimes, of course, there are time lags. Expectations attached to the concepts of gender and family may change more quickly than the physical form, making it difficult to enact new expectations without considerable hardship (n.d.:18).
Thus human built environments are not neutral. They are the physical manifestation of a patriarchal, capitalist socie ty's expectations of what types of activities take place where, when, and by whom (Bowlby and McDowell, n.d.; Boys,
1984; Coser, 1982; Freeman, 1981; Markusen, 1981; Wekerle,
1981).5x But societal norms and values are so deeply ingrained and relatively ubiquitous that their reflection in human built environments often goes unacknowledged (Ley,
1977; Mackenzie, 1987).52 However built environments are
51 This is not to say that the built environment controls women's lives, but they do work in partial conjunction with many other social, political and economic relations to keep people in their "place" and do physically sym bolize what that "place" is (Coser, 1982).
52 Accordingly, a sensitive reading of the built environ ment can generate important insights into the social processes and assumptions that they reflect and that they help to perpetuate. As Hillier and Hanson (1984) 104 not merely the concrete manifestations of past and present ideas about gender and other social relations; they also provide the conditions for the reproduction and hopefully the transformation of gender roles and relations (Boys,
1984; Giddens, 1979, 1985; MacKenzie, 1987; Sayer, 1984,
1985).53 Thus architects, developers, business-people and planners (most of whom men) have been criticised for
remark: "The ordering of space in buildings is really about the ordering of relations between people. Because this is so, society enters into the very nature and form of buildings. Buildings are social objects in a way quite incomparable to other artefacts, in that society is involved at the deepest level in determining their forms as objects. Architecture is not a 'social art' simply because buildings are important visual symbols of society, but also because through the way in which buildings, individually and collectively, create and order space, we are able to recognize society: that it exists, and has a certain form" (1984:33).
53 It is also important to realize that women are not pas sive victims of the human built environment. Far from it, women are actively changing the human built environ ment. For instance Stamp (1980) describes how some women are altering their neighborhoods so that they will be more supportive of the changing requirements of their life-styles and facilitate self-development and self growth. Thus Stamp declares: "So while we must still occupy the built environment of the city in its present form. we are beginning to break down its isolating bar riers within our groups and within neighborhoods, and assert that its function will no longer remain as it was. The old form will no longer have the power to cripple us by stifling our growth, and this may be the most powerful step toward reshaping the form itself and breaking down the barriers of segregation between 'us' and 'the other'". A second example is that of lesbians who have been creating places in cities which are "safe spaces" which provide a supportive environment relative ly free of homophobic prejudice. While such spaces are limited (although they are not necessarily distinguisha ble) they are often located in university or "bohemian" neighborhoods (Beyer, 1985; Ettorre, 1978; Holcomb, 1986). 105 designing dwellings and cities based on patriarchal assump tions about "appropriate" gender roles (Boys, 1984; Hayden,
1981; Holcomb, 1984a; Keller 1981; MacKenzie, 1983, 1984,
1987; Stimpson, Dixler, Nelson and Yatrakis, 1981; Wekerle,
Peterson and Morley, 1980; Matrix, 1984).54 As Wekerle
(1984) puts it:
Cities are still planned by men for men. While the lives of women have changed radically, the urban environment in which they live has not. A good deal of urban development has shown a con ventional bias favoring independent nuclear fami lies and planning for housing and communities reflects an image of the family where the woman is a full-time home-maker and caretaker of the home and the man is a full-time wage earner. This polarization of male and female roles is mirrored in the high degree of segregation between work-place and residence found in the North American city. Recent studies document the inordinately high costs to women of transporta tion systems, neighborhood planning, and urban housing markets which reinforce traditional roles and limit women's access to new opportunities (Wekerle, 1984:11-12).
So today's cities of largely residential suburbs segregated from areas of production are, in part, based on assumptions about gender roles and gender relations and subsequent social practices.55 Clearly the whole notion of suburban
54 Although attention here will concentrate on the human built environment of the city, it should also be noted that feminists have argued that most housing isolates family units ensuring the minimum privacy for its mem bers and thus maximizes the domestic work of individual women. Furthermore large, new houses and the internal layout of those houses — large glass windows, open plan settings and galley kitchen — add to the visibility of housework, enforcing high standard of cleanliness and neatness (McDowell, 1983).
55 Of course the perceived independence of these spheres is 106 residencies in far flung parts of the city segregated from sites of production, is implicitly grounded in the social construction of the nuclear family with a full-time female homemaker (McDowell, 1983; MacKenzie, 1983, 1984; MacKenzie and Rose, 1984).56 State intervention, zoning regulations and urban planning have helped to create and perpetuate suburban single-family residential areas-. Gender inequali ties and notions of the nuclear family were implicitly embraced, along with the "suburban mentality" of the arca dian life-style where the "home is a haven" with an isolat ed full-time wife and mother staying at home, segregated from the distant sphere of production (Banner, Berhelde and
Greckel, 1982; Bowlby, Foord, McDowell and Momsen, 1983;
Graff, 1982; McDowell, 1983; MacKenzie and Rose, 1983; Net- ter and Price, 1983; Wagner, 1984; Winter and Morris,
1982). The expansion of suburban developments in the
1950's increased the length of work-trips for men and increased the isolation women felt from being removed from their support networks in the city. If women wanted to
a false one as, in reality, they are mutually dependent and reinforcing.
56 Largely excluded from this life-style have been many lower income families, particularly if they were non white. As many of the women in these families were in the paid labor force, some feminists believe that the "leisured" life of the middle class suburban women was, in part, dependent on the labor of their working class urban sisters. However, the history of "suburbaniza tion" should not be seen only in terms of an attempt to establish an "ideal family" in a sphere separate from production. 107 take on waged employment it was made difficult by the lack of locally available job opportunities, at least in the early stages of suburbanization, and by inadequate trans portation services (Bowlby and McDowell, n.d.; Rothblatt,
Garr, and Sprague, 1979; Wekerle, Peterson and Morley,
1980).
The history of American cities, therefore, can be viewed as one of increasing spatial, temporal and functional sepa ration of the home from the work-place. The pre-industrial city characterized by the unity of home and work, was grad ually replaced by cities of patchwork areas of specialized reproduction -- residential suburbs — and specialized pro duction in the 1950's (Hayford, 1974; MacKenzie 1983, 1984;
Westwood, 1982). Indeed the 1950's is considered to be a benchmark decade for the relationship of women to housing, in that the new full-time "house-wife" came into her own.57
Almost regardless of class, women became all purpose "high- value low-cost" housewives (Oakley, 1974), with responsi bilities for the upkeep of their new single-family dwelling as a safe haven for the emotional well-being of their fami ly, for reproducing the labor force, and for being active consumers of mass-produced goods and services (Miller,
57 Clearly this is a generalization. Forms of the "house wife" probably existed prior to the 1950's, but they have become relatively "invisible" to us (except, per haps through novels). However the significance of the 1950's is that it was the decade in which the house-wife role gained mass popular appeal. 108
1983; Ravetz, 1984). Indeed, as noted in Chapter Two, home-ownership and suburbanization generated a new life style based on the private consumption of consumer- durables, and many suburban owner-occupiers rely on the appreciation of house values to increase their own purchas ing power (McDowell, 1983). However, as most suburban residential neighborhoods were planned and built only to facilitate one role for women, that of full-time wife and mother, MacKenzie argues that:
This always tenuous bundle began to unravel as women took on (multiple) roles, and changed from being ("naturally") passive, nurturing creatures to dynamic and efficient actors ... She moved from one place to the other every day, tying pub lic and private, home and work, together. She did so with the help of radial bus services, inadequate child-care, shops that closed at 5 pm and a family and home which was in need of con stant attention and complex planning. In many ways, a city which had worked for women who were full-time suburban domestic workers, no longer worked for women who were partially responsible for financing the suburban home as well as main taining it (1983: 17).
3.3.3 Residential Location and Gender
Women in two-earner households tend to be more con strained than other women because residential choices must also depend on their partner's needs (Gordon and Molho,
1985; Madden, 1981). As married women tend to have shorter job tenures than their partners, residential location is often selected on the basis of the specific job location of the man, and the spatial distribution of potential job 109 locations for the woman (Fagnani, 1983; Madden, 1981).58
Furthermore a number of studies have found that while men favor suburban residencies (they enjoy being able to retreat to a "relaxed" life which offers many outdoor activities), women are much more ambivalent. They value the perceived safe and homogeneous environment of the sub urbs, but also find them isolating and boring, as opposed to the city which is seen as stimulating and where they can enjoy numerous amenities and services.
Women in paid employment who live alone (or at least without a husband or children) are more likely to locate in more centralized residential areas than other forms of households do (Freeman, 1981; Kaniss and Robins 1974; Mad den, 1981). Indeed there is a definite concentration of female headed households and single adults and childless adults in the central city: Holcomb (1984a) found that cities have consistently higher gender ratios than either suburbs or rural areas. Central cities, especially when compared to single family suburbs, have more varied housing and services, and closer facilities. Such an environment is potentially more supportive of women in general, but particularly those women in "non-traditional" households and mothers who are waged laborers (Holcomb, 1984a, 1984b;
5 8 However it should be pointed out that the more committed a women feels about her job/career the more important will her job location be in residential location deci sions . 110
Rose, 1984; Wekerle, 1984).59
3.4 SPATIAL ENTRAPMENT AND FEMALE CLERICAL EMPLOYMENT.
One body of literature in geography which explores the spatial implications of "captive" labor forces is that which deals with spatial divisions of labor, including that focusing on the suburbanization of office employment. The thrust of this literature is that the standardization and routinization of the production process have provided the necessary conditions for the fragmentation of the produc tion process into functionally and spatially separate units of production. The significance of the existence of sepa rable units is that they may be (separately) located wher ever unevenly distributed locational conditions, such as wage levels, labor supplies, and labor union organization, are best suited to that specific unit of the overall pro duction process. Thus many firms have been reorganized so
5 9 While these women presumably enjoy this proximity to jobs, services, and activities (Gerson, 1982; Roistacher and Young, 1980), the high gender ratios are also reflective of the high concentration of female headed households in the city. Indeed 41% such households (which are often more centralized due to their lower incomes and inability to pay the extra housing and transportation costs of less centralized locations) located in the central city, compared to only 22.9% of "traditional" households (Gerson, 1983; Holcomb, 1984b). In the Columbus metropolitan area, Cook and Rudd (1984) have found that female headed households tend to be con centrated in the downtown are and in some of the older suburbs. This could be due to a mismatch of new housing development and job opportunities, forcing some women to look outside the immediate neighborhood for paid employ ment . Ill that research, development, and managerial functions are located in one location, while direct production often takes place in branch plants in the "periphery" (Clark,
1984; Dunford, 1979; Hall, 1985; Massey, 1983, 1984, 1985,
1987; Taylor and Thrift, 1983; Toft Jensen, Hansen, and
Serin, 1983).
The role of women as a spatially captive pool of labor has played a significant part in spatial restructuring, and numerous spatial divisions of labor and locality studies have explored this linkage. The vast majority of these have focused on the regional division of labor in Britain.
Favorite regions for analysis have been those which have recently experienced a rapid growth in female paid employ ment after a long history of economic domination by the male-employing coal-mining, ship-building, steel and heavy engineering industries. Emphasis is placed on the function of pre-existing gender roles, relations and ideology in helping to shape the present spatial form of paid employ ment (see for example: Hudson, 1980; Lewis, 1984; Lewis and
Foord, 1984; McDowell and Massey, 1984; Massey, 1983, 1984;
Murgatroyd and Urry, 1985; Walby, 1985; Williamson, 1982;
Winckler, 1985).60 As Bowlby, Foord and McDowell (1986) concisely summarize:
60 Some studies have considered other (predominantly rural) regions where there has also been a rapid increase in the paid employment of women (see for example: McDowell and Massey, 1984). 112
...they tend to adopt an argument which runs as follows: through local and historical conditions of production the kind of work which became available was for men. It was also heavy, dirty and on a shift basis. This in turn resulted in heavy and demanding domestic labor for women which prevented them from working for money out side the home. This, in turn, meant that women constituted an inexperienced labor force. Subse quently this pool of inexperienced labor which was also "cheap" attracted new kinds of "light", labor-intensive, female-employing industry. This kind of account is often paralleled by descrip tions of the male-dominated community life and politics which have emerged in such areas (1986:328).
However it is also possible to detect a new geography of economic activity at the metropolitan scale. There has been a loss of heavy manufacturing jobs in the inner city, and an expansion of job opportunities in light manufacturing and the service sector, particularly in suburban locations.
As McDowell (1983) noted:
... the pool of married women on ...suburban estates proved an attractive and flexible source of labor for the light-assembly industries that also began to decentralize and expand in the sub urbs .
And in clerical work, developments in the automation and routinization of office employment and in communications technology have allowed more routine functions, such as payroll, billing, claims processing, to be separated from those non-routine aspects of office work requiring face-to- face contact with clients (Cervero, 1986; Crompton and
Reid, 1982; Feldberg and Glenn, 1988; Glenn and Feldberg,
1982, 1984; Massey, 1984; Walby, 1985, 1986; Winckler, 113
1985). Given the conventional view that women cope with their multiple roles by confining their activities to a limited geographical area, it is suggested that routine aspects of office employment can be relocated to the sub urbs, specifically to take advantage not only of lower rents and lower crime rates, but also of the availability of this supposedly spatially entrapped pool of female labor
(Bruegel, 1979; Lewis, 1984; MacKenzie and Rose, 1983; Nel son, 1986a, 1986b; Saegart, 1981). As Kaniss and Robins
(1974) note:
Lack of mobility for the worker, according to traditional economic theory, can result in depressed wage rates because the worker is unable to migrate to locations of higher-paying employ ment and is forced to accept a lower-paying job. Women, as we have seen, are forced to search for work near their homes. Thus women become a cap tive labor force in their area of residence, obliged to accept lower-paying jobs or part-time employment as a substitute for full-time work ... (and) ... many firms have taken advantage of this situation by deliberately choosing locations which are accessible to this pool (1974: 66; emphasis added).
The suburbanization of office-work is a well documented trend for the United States and other advanced capitalist countries (see for example: Alexander 1979; Armstrong,
1972; Cervero, 1986; Daniels 1974, 1979, 1982, 1985).
Although this literature usually concentrates on very large metropolitan areas such as New York City, Seattle, London and Toronto, Selwood (1985) has identified this trend in the Columbus metropolitan area. He found that while both 114 routine and non-routine office employment have been decen tralizing since 1970, the suburbanization of routine office work was occurring at a faster rate.61
Although usually not couched in terms of the spatial divisions of labor literature, a number of authors have examined the relationship between the location of routine office employment in the suburbs and the geography of the supply of female clerical workers (for examples Alexander,
1979; Baran, 1985, 1986; Nelson, 1986a, 1986b; Sassen-Koob,
1984; White, 1978; 9-to-5, 1985). For instance, in a study by 9-to-5 (1985) it was suggested that:
Suburbanization ... of office employment might occur for three reasons: the growing technologic al capability of employers to link up decental- ized work sites through telecommunications, the raising cost of office space in the central cit ies and a preference by management to hire white women workers (9-to-5, 1985:9-14).
Nelson (1986a) has attempted to explain the relationship in terms of the characteristics of female clerical labor supplies in various residential areas of the San Francisco
Bay area. She describes how until the Second World War most offices were downtown, employing mainly young, white, native-born, relatively well-educated women who worked
61 In an earlier study, Selwood (1982) employed density gradients to examine the decentralization of different sectors of office work in the Columbus metropolitan area for 1964, 1970 and 1980. He found that the various sec tors usually exhibited declining density gradients over time, although there was some variation: for instance, government remained concentrated downtown, while finance, insurance and real estate showed considerable decentralization. 115 until they got married. Since the Second World War employ ment patterns have changed quite dramatically. Downtown- office employers have become increasingly reliant upon minorities for their clerical labor force. Nelson suggests that these employment shifts have been partly due to white flight to the suburbs and the tendency for many young women to remain in education; both of these have reduced the sup ply of young, white, single women clerical workers in cen tral city locations.62 Indeed, according to Nelson:
What has become scarce in U.S. central cities is the relatively well-educated female worker who is content to stay with low paid, dead-end clerical work (1986: 157).
Partly this has been a matter of low educational attainment among minority groups; but more significant is the rela tively large number of single mothers among downtown female workers. Many women residing downtown are actually the chief economic supporters of their families, working to provide basic family and household needs.63 Thus the wages earned by many inner city women cannot be treated as a sec ondary source of family income. Given this, it is not sur prising that employers have found an increase in militancy among downtown female workers and a substantial labor turn-
62 Note that this also describes the type of woman conven tionally viewed as less (if at all) spatially entrapped.
63 A second group of women also live in the downtown area -- middle-class, primary earners. However they live downtown largely out of choice rather than necessity, in order to enjoy its gentrified ambience (Markusen, 1981). 116 over as these women attempt to stabilize and improve their economic positions. As a result, Nelson argues that:
Suburban areas of new, single-family housing are now the best source of (desirable clerical) female labor, due to their class structure and related household structure ... Relatively edu cated women who have deferred career plans for domestic responsibilities are now most commonly found in "starter home" suburbs, because families with young children and headed by husbands are prevalent, while social mechanisms for sharing child-rearing are uncommon ... Suburban communi ties aimed at young, growing families are thus the best place for employers to find women whose primary responsibilities lie in the household, but who are eager for alternative employment that can fit around domestic duties (1986:157-158).
Nelson is clearly drawing on the established (but
increasingly less realistic) view in employment studies
that suburban women have a negative attachment to the paid
labor force, due to their domestic obligations and ability
to perform domestic labor in return for their husband's
economic support (see for example: Cain, 1966; Ginzberg,
1968; Schultz, 1972). Nelson argues that suburban women
possess the characteristics of the "desired clerical labor
supply (as they are) relatively educated women whose heavy
child-rearing responsibilities restrict their career mobil
ity" (1986b: 5). At the same time they possess the language
skills that have become so important in customer interfac
ing. In addition their "class structure and related house
hold (home-ownership) structure" enhance employee stability
and productivity, while on-the-job resistance or feelings 117 of militancy tend to be limited. As a consequence accord ing to Nelson:
To the managers of the offices employing large numbers of low-wage clerical workers, a female labor supply associated with areas of growing single-family housing represents a significant lowering of labor costs through reduced turnover, lowered training time, increased productivity, a longer working day, and a reduced chance of unionization. And two important elements of this cost-saving labor supply, its cheapness and attachment to home responsibilities, forfend long commuting distances; therefore firms must locate offices nearby to achieve the potential savings (1986: 165-166).
Clearly there are two important underlying assumptions to Nelson's argument. First, suburban married women with young children, spatially entrapped in their neighborhoods of residence, form a captive, low-wage, well-educated pool of labor. For instance, she remarks that one of the manag ers she interviewed:
indicated they were limited to the local labor market for clerical labor ... since most clerical workers are restricted to short journeys to work. Like other industries dependent on female labor, back offices must be located very near their intended labor supply: it has been demonstrated in numerous empirical studies that women who work outside the home have shorter journeys to work, on average, than do men (1986: 152).
Second, there is an implicit acceptance of particular types of suburban women needing paid employment. Specifically these women are those with "starter homes" (that is newly constructed, owner occupied housing) in which they live with their husbands and young children. Essentially Nelson 118 argues that due to their household composition, particular suburban women have a need for paid employment, but due to their domestic obligations they can only support short journeys-to-work. This, in addition to their forming a
"preferred clerical labor supply", means that firms relo cate to the suburbs in order to enjoy "potential savings" accruing from this captive labor force.
3.5. EVALUATING SPATIAL ENTRAPMENT: MOBILITY STUDIES
Taking Nelson's lead, trip behavior studies provide one possible tool for evaluating the spatial entrapment thesis.
Most transportation studies examining gender differentials in work-trip patterns do (as Nelson remarks) draw similar conclusions: regardless of the place or time period exam ined, women generally have a shorter mean journey-to-work than men, both in terms of time and distance (Allen and
Edmonds, 1979; Ericksen, 1977; Hanson and Hanson, 1980,
1981; Howe and O'Conner, 1982; Madden, 1981; Fagnani, 1983;
Hanson and Johnston, 1985).64 Table 15 shows65 that recent
64 Journey-to-work time is a function of the distance tra velled, the mode of transportation used and congestion levels. Different results could be produced using travel time rather than distance. For example, it is plausible that women's greater reliance on public transportation might be reflected in shorter distances to work, but longer journey times to work.
6 5 The authors featured in this table analysed data at var ying spatial levels. For instance Madden uses national data (the Panel Survey of Income Dynamics), while Hanson and Johnston employ data generated for Baltimore. 119 journey-to-work studies estimate that the mean commute for women is between approximately 7 and 9 miles or 26 to 37 minutes, while for men it is between 8 and 13 miles or 29 to 53 minutes.
Table 15
Summary of Journey-to-Work studies: mean miles (and minutes)
AUTHOR DATE WOMENMEN U.S. Department of Commerce 1979 7.00 11.00 Cichocki 1980 9.00 13.00 Madden 1981 8.25 11.25 Fox 1983 (37) (53) Brooker-Gross and Marraffa 1985 8.70 8.00 Hanson and Johnston 1985 5.50 7.80 (25.7) (28.8) Rutherford and Wekerle 1988 6.80 8.80
Studies in North America, however, indicate that both women's and men's worktrips have lengthened over time (Ber ry and Kasarda, 1977; Rutherford and Wekerle, 1988; Yeates and Garner, 1980). In their study of metropolitan Toronto,-
Rutherford and Wekerle (1988) found that between 1971 and
1981 women's average commute increased from 5.1 miles to
6.8 miles, while men's increased from 7.5 miles to 8.8 miles. They further note that not only had the proportion of the labor force working close to home fallen over the same time period but that the proportion commuting long distances had increased. Whereas 64% of women and 46% of men worked within four miles of home in 1971, only 51% of women and 38% of men worked within five miles of their 120 homes by 1981. Looking at longer commutes they found that the rate of increase for the proportion of women commuting more than thirteen miles (6% in 1971 and 13% in 1981) was more rapid than that of men (15% in 1971 and 22% in 1981).
Numerous studies have also examined women's travel pat terns from the viewpoint of the constraints women face when they combine their various roles. Theoretically one would expect that the greater a person's household responsibili ties, the shorter that person's work-trip would be. That women are traditionally responsible for the home and chil dren is one of the major explanations which has been for warded as to why women have shorter work-trips than men
(Ericksen, 1977; Fagnani, 1983; Madden, 1981; Madden and
White, 1980). Several authors suggest that while the ages and number of children affect the probability of a woman being in the paid labor force (as well as whether she works full- or part-time), the same does not apply to men (Erick sen, 1977; Fagnani, 1983; Hanson and Hanson, 1980, 1981;
Smith, 1979; U.S. Department of Labor, 1980). However the results of empirical studies on the relation between the journey-to-work and family responsibilities do not corrobo rate each other. Thus Fagnani (1983) found that the greater the number of children that a women had, the closer to home she worked. Madden (1981) combined both the number and age of children aged less than three in a household along with 121 other characteristics of the spouse and the family to anal yse the contribution of household responsibilities to women's shorter work-trips. She concluded that gender dif ferences in the domestic division of labor were more impor tant than income or labor force characteristics in account ing for women's shorter work-trips.
Regarding marital status, Ericksen (1977) found that married women have the shortest commuting time and single women have the longest commutes. On the other hand,
Brooker-Gross and Marraffa (1985), Madden (1981) and Hanson and Johnston (1985) found married women to have longer journey-to-work distances than unmarried women. Interest ingly, when Hanson and Johnston (1985) evaluated the impact of household responsibilities (based on the presence/ absence of pre-school and school-aged children and the presence/absence of other working/non-working adults in the household) on women's commutes to work they did not find an inverse relationship between work-trip length and the level of household responsibilities. Not only did they find that single women ("theoretically1' the least constrained) had the shortest work-trips, but they also found that neither the presence of children nor another adult reduced women's work-trip length.66 In short journey-to-work studies con
66 They did, however, find that women did have shorter worktrips than men, except for single women and men whose trips were not significantly different from one another. This provides further support for Madden's (1981) hypothesis that intra-household gender roles 122 sidering women's domestic roles have produced conflicting results and no definitive conclusion can be drawn.67
3.6 SUMMARY
This Chapter reviewed literature dealing with questions of space as they relate to women's paid employment in gen eral and to clerical work in particular. The first section provided a conceptualization of spatial entrapment, drawing upon time-space budgeting concepts. Then attention turned
to a consideration of how historically received gender
roles intensify the problems of time-space budgeting,
implications. The next section drew upon the work of Kris
affect, in part, women's work-trips but that the pres ence of children does not.
67 Given the focus of this dissertation it is noteworthy that the few mobility studies considering gender and occupation are equally inconclusive. When Hanson and Johnston (1985) examined journey-to-work patterns by gender and occupation they did find that clerical work ers had among the smallest commutes (5.3 miles or 24.8 minutes). Furthermore they found that working in a female-dominated occupation, regardless of gender, was associated with shorter work-trips than working in male- dominated occupations. When Cubukgil and Miller (1982) looked at the work-trip patterns of six different occu pation categories in Toronto, they also found that cler ical workers had among the most localized commutes. Conversely, Wheeler (1967) found that female clerical workers had a commute of 3.46 miles, which in his study was one of the longest commutes, and Fagnani (1983) found high commuting rates (women working outside their district of residence) for clerical workers in her study of Ile-de-France. In short, and somewhat contrary to Nelson's assumption that clerical workers have short commutes, there have been varied and contradictory empirical results for clerical workers in journey-to- work studies considering occupation. 123 tin Nelson to explore the concept of spatial entrapment in light of the changing geography of the supply of female clerical workers and it's implications for office location.
Finally journey-to-work studies were reviewed in an initial evaluation of the spatial entrapment thesis. CHAPTER IV
METHODOLOGY
4.1 INTRODUCTION
There is no single best method in science. Rather appropriate method depends, in the first place, on the nature of the object of inquiry: in particular on whether it is a natural or social object. And secondly upon the purpose of the inquiry: is it for explanation, calculating quantitative results or prediction? This chapter explores the implications of these questions for social science research. This is approached first by examining contrast ing approaches to theory. Two contrasting methods of research are then described. The chapter closes with a discussion of the methodologies and data selected for this particular dissertation.
4.2 CONTRASTING APPROACHES TO THEORY
This section compares and evaluates two competing theo ries of how research should proceed: the ordering-framework conception of theory versus conceptualization. The ordering-framework conception of theory involves the search
- 124 - 125 for empirical regularities and universal laws of human activity with the aim of discovering order in the world.
Conversely theories such as realism seek explanations of why things happen the way the do. This in turn depends upon careful conceptualization of those structures and cau sal properties of objects which enable them to produce or suffer particular types of change.
The ordering-framework conception of theory is one in which observations (data) are slotted into a "model" large ly with the aim of successfully identifying enduring empir ical regularities, predicting unknowns, and identifying universal laws of human behavior. The description and con ceptualization of the categories into which observations are slotted are rarely questioned. When they are it is usually with regard to taxonomic adequacy rather than whether the category is a true and rational abstraction (as opposed to a chaotic or bad abstractions where the unrelat ed are combined or the indivisible are divided).68 This is largely because the characteristics of the objects studied are taken as given by the objects themselves. They are seen as things-in-themselves with their own self-defining
68 In part the ordering-framework conception of theory is based on the adoption of the methods of the natural sci ences. Natural scientists seemingly (to the social sci entists) obtain spatially and temporally universal laws of physical behavior, which have enabled them to fairly successfully predict the outcome of the interaction of their objects of study. Hence in an attempt to predict, some human geographers have employed similar methods to identify universal laws of human behavior. 126 naturally occurring characteristics, rather than as defined by their relations to other objects. Accordingly, although it is not explicitly understood in the ordering-framework conception of theory, all relations are external and con tingent (as opposed to internal and necessary), apprehendi- ble through observation as "facts": distinct, discrete tangible objects and events. These "facts" are given to the senses in that people's perceptions of them are consid ered to be conceptually unmediated and unaffected by any pre-existing theories, concepts or ideas (including lan guage). In short the object of investigation is considered to be separable from the investigator (or subject) and is theory-neutral.
There are a number of problems with this approach.
First, knowledge of the world is never conceptually unmedi ated. People selectively perceive the world through a set of shared intersubjective meanings. These "facts" are, therefore, (pre)conceptualized by virtue of the shared cul tural concepts, theories and symbolic systems with which people act. So observation is actually theory-laden.
Indeed such a selective perceptual framework is a necessary precondition for all human activity (academic or other wise ) .
Second, the regularities uncovered in human geography
tend to be true only for short segments of time and for 127 specific places: generally, and in consequence, the success rate for prediction has not been very high.69 This is because the production of empirical regularities, necessary for successful prediction, requires two preconditions.
First, the internal structure of the object must always be consistent (Bhaskar refers to this as "intrinsic conditions for closure"); second the external conditions must never vary ("extrinsic conditions for closure"). If a system satisfies both of these conditions it is a closed system; if not, it is an open system. Successful prediction is only possible in closed systems and while some natural sci entists deal with closed systems (the consistent outcomes of which can be successfully predicted), social scientists do not. Their objects of inquiry do not meet the condi tions for closure, because people do not mechanistically react to, "stimuli". Indeed they have the distinct quality of being able to learn, understand, respond to, possibly in novel ways, and transform "stimuli". So outcomes are not always the same, thus making prediction a highly hazardous operation.7 0
69 The usual explanation given for poor prediction in the social sciences is that they are "immature" and that true order will only be achieved if research continues on a "scientific" path.
7 0 Sayer remarks that this aspect of the social sciences should not be viewed as a weakness. Instead he points out that as academics are both the subject and object of social systems they actually have an advantage over nat ural scientists. As human beings they have an internal access to understanding their objects of study. Through 128
Under the realist approach to research, on the other hand, the rational abstraction of structures takes priority over the construction of ordering frameworks. This is directly linked to the realist’s conception of causality, which is understood as the necessary way of acting (or cau sal mechanisms) of an object by virtue of its structure.
These mechanisms or particular ways of acting consist of both causal powers -- people have the causal power to work, speak and think; and of causal liabilities -- people can suffer from unemployment, from not eating, and from peer pressure (Sayer, 1984). These powers and liabilities exist even when they are not being exercised or suffered, and when they are exercised their resultant effect is dependent upon the contingent conditions under which they were exer cised or suffered. In other words these mechanisms are distinct from the effects that.they generate (Sayer, 1984).
As Bhaskar (1975) remarks:
The citation of a (realist) law presupposes a claim about the activity of some mechanism but not the conditions under which the mechanism operates and hence not about the results of its activity (1975:95).
In short one cannot simply "read off" effects from know ing the existence of some mechanism: the same causal prop erties could produce completely different concrete out comes.71 Conversely, that a regularity exists and may be
shared meanings and language they can, for example, interview people and conduct thought experiments. 129 predicted does not provide an understanding of what pro duced this regularity or why it occurred, it may have been caused by any number of causal properties. concrete out comes. As Sayer (1984) says:
Contrary to popular myth (derived from the asso ciation of causality with regularity) what causes a (social) event has nothing to do with the num ber of times it has been observed to occur and nothing to do with whether we happen to be able to predict it." (1984:100)
4.3 CONTRASTING APPROACHES TO METHODOLOGY
Given the assumptions of the ordering-framework concep tion of theory (the implicit notion of externally related objects and unvarying outcomes resulting from the constant conjuction of the same sets of objects) the use of quanti tative methods to discover empirical regularities appears to be perfectly adequate. The problem with such an epis- temology is that the objects of inquiry which most human geographers deal with (people and the outcomes of social activity), are, in fact both internally and externally
related.
It is for this reason that realists concentrate so much attention on the conceptualization and abstraction of their objects of inquiry. For if the discovery of causality is
the aim of research then the objects of study must be care
71 There may be other unacknowledged mechanisms involved in producing the object of inquiry. Careful observation and description are needed to avoid this. 130 fully and rationally abstracted and conceptualized. So, in contrast to the ordering-framework conception of theory, the larger part of the realist investigation is spent uncovering causal mechanisms capable of producing the object of inquiry. For instance, in the discussion of the paid employment of women, we should ask ourselves what is it about suburban women (and women in general) which makes them cheaper to employ (and oftentimes more productive) than men? Is it a natural/biological result of their gen der or is it a result of certain aspects of patriarchal hierarchy? Is spatial entrapment necessarily a result of being a women and/or mother and living in the suburbs, or is this just one possible outcome? Essentially then, when it comes to method, the realist tries to take account of the ontological aspects of the object of investigation: ontology (object studied) should guide epistemology (method used to study the object). It can be defensibly argued, however, that most geographers have allowed epistemology to precede ontology.7 2
Methods which are concerned with discovering general patterns and common properties among a representative sam ple of the population are described as those of extensive
72 The realist's ontology includes the existence of a "real world" outside of our present perceptions and experi ences -- objects and events still exist whether tangible or not. Just because at one time Australia and North America had not been discovered did not negate their existence. 131 research. In intensive research, on the other hand, the primary research questions are concerned with how causal properties are manifest in a particular case or number of cases. As Table 16 indicates the distinction between the two types of research is more than simply "breadth versus depth". They ask two different sorts of research question, use different methods and techniques, and have different definitions of their objects of inquiry. For instance, the typical methods employed in extensive research include descriptive and inferential statistical analyses and large- scale standardized questionnaires of "representative" sam ples, or even of whole populations. Intensive research, on the other hand, uses largely qualitative analyses including less formal, less standardized, interactive, in depth interviews, participant observations and ethnographies.
Both forms of research are important but perform differ ent functions — generally, one is descriptive and synop tic, while the other is explanatory (Sayer, 1984; Sayer and Morgan, 1985). Both forms of research also have their disadvantages. For instance, intensive research does not give any indication of how representative the results pro duced by a particular causal process are. On the other hand, extensive research might fail to indicate which pro cess produces the patterns observed.73
73 For instance, when quantitative methods are used to study human activity it should be recognised that there are many intangible aspects of the world which are not 132
Extensive research might throw light on the questions of how and why the world is the way it is. Indeed, by using extensive research a realist might successfully uncover some necessary causal relation producing some regularity, but she/he does not expect to find enduring empirical regu larities, given the focus on humans and social activity.
Essentially extensive research is only suggestive when it comes to identifying causes and causal structures. It is this problem which intensive research aims to resolve.
Accordingly considerable emphasis is placed on verstehen. that is an interpretation of the actors own understanding of the circumstances of their actions. As such intensive research specifically attempts to unearth: (i) the causal
(social) structures in which people are implicated and which give them the powers or needs to do things or the liabilities to suffer in certain ways (powers to change, and susceptibility to be changed); and (ii) the contingent circumstances which result in the exercise (or suffering) of those powers (and liabilities) and which give empirical expression to them. Thus in intensive research the observed cases can be examined through, for instance, the use of ethnographies and interactive interviews (Sayer,
1984: 219). Basically the aim of intensive research is to achieve a greater understanding of what particular types of
quantifiable. The unrestrained and poorly conceptual ized use of quantitative methods may result in the neg lect of these unquantifiable aspects. 133 necessary relations or causal structures exist and bear upon the research question. 134
Table 16
A Summary of Intensive and Extensive Research
INTENSIVE EXTENSIVE Research How does a process What are the regularit- Question work in a particu- common patterns, or small number of distinguishing features cases? What produces of a population? How a certain change? widely are certain char What did the agents acteristics or processes agents actually do? distributed or repres ented?
Type of Causal explanation Descriptive "representa account of the production of tive" generalizations, produced certain objects or lacking in explanatory events though not penetration. necessarily repre sentative ones.
Typical Study of individual Large-scale survey of Methods agents in their cau population or represen sal contexts, inter tative sample, formal active interviews, questionnaires, stand ethnography. Quali ardized interviews. tative analysis. Statistical analysis.
Are the Actual concrete Although representative results patterns and cont of a whole population, general- ingent relations are they are unlikely to be izable? unlikey to be "repr- generalizable to other sentative" or "aver populations at different age", or generaliz- times and places. able. Necessary relations discovered will exist everywhere their relata are pre sent, e.g. causal powers of objects are generalizable to other contexts as they are necessary features of these objects.
Disad Problems of repre- Lack of explanatory power. vantages sentatativeness. Problems of ecological fallacy in making infer ences about individuals.
Source: Sayer and Morgan (1985:156) 135
4♦4 METHODOLOGY EMPLOYED IN THIS DISSERTATION
Informed by the above discussion, this section provides a description of the methodology and data employed in this dissertation. The first section concentrates on the exten sive research, while the second section presents the type of intensive research used.
4.4.1 Extensive Research
The methods employed in the extensive research include simple statistics, shares, location quotients and multiple regression analysis. A location quotient measure the extent to which a particular Census tract has more or less than its proportional share of female clerical employees, where "proportional” shares are constructed on the basis of shares of the female population in toto. Thus the index is constructed by dividing the percent of Franklin County's female clerical employees falling in a particular tract by the percent of Franklin County's women of paid labor force age in that same tract. A value of 1.0 indicates that a
Census tract's share of female clerical workers is propor tional to its share of women of labor force age. Values over 1.0 connote an overrepresentation of female clerical workers, while values under 1.0 indicate an underrepresen tation. 136
4.4.2 Intensive Research
Formal, standardised interviews (favored in extensive research) were not employed in the intensive research for this dissertation. This is because the rationale behind them suggests that by asking each respondent the same ques tion under "quasi-experimental" conditions the observer's bias is reduced and so controlled comparisons are possible.
As Sayer and Morgan (1985) explain:
Such ideas sacrifice explanatory penetration in the name of "representativeness" and "getting a large enough sample". Extreme standardization which disregards the different types of respon dent can in fact make comparisons rather meaning less, because they fail to register the fact that different questions can have a vastly different significance for different respondents (1985:156) .
Instead unstructured, interactive interviews are employed.
"Unstructured", of course, is a relative term. An inter view guide providing a checklists of items touched upon in the course of each interview is shown in Appendix A and B.
Due to the double hermeneutic in the social sciences and the intersubjectivity of human actions, it is inevitable that interviews are, in fact, interactive. Instead of attempting to minimize interaction (in order to minimize observer-bias) which may lead to an awkward "pseudo conversation", interaction was consciously used to maximize the information gleaned. This did not mean, however, that the interviewer tried to influence the interviewees, but it 137 did involve being flexible in question-asking and shifting the emphases in the interview in accordance with what the interviewee was able to talk about (Finch, 1984; Oakley,
1981; Sayer and Morgan, 1985; Sayer, 1984).
Unstructured, interactive interviews are also very much in keeping with the goals and politics of feminist analy sis, and for at least two reasons. First, they help avoid a hierarchical relationship between the interviewer and interviewee. Second, their interactive (and non- hierarchical) aspects allow a more collaborative approach to research by engaging both interviewer and interviewee in a joint project, indeed "the formulation of a relationship between interviewer and interviewee (is) an important ele ment in achieving the quality of the information required." (Rapoport and Rapoport, 1976:31).
4.5 SUMMARY
This chapter reviewed two contrasting approaches to theory: the ordering-framework conception of theory and realism grounded in conceptualization. This was followed by a discussion of two different forms of research method ology: extensive (such as statistical analyses and formal questionnaires) and intensive (such as interactive, unstructured interviews and participant observation)
research. In the final section the methodology and data 138
(for both extensive and intensive research) employed in this dissertation was presented. The results are reported in the next two chapters. CHAPTER V
WOMEN'S MULTIPLE ROLES AND SPACE: A QUANTITATIVE
ANALYSIS
5.1 INTRODUCTION
This Chapter examines Kristin Nelson's (1986) explana tion for the suburbanization of female clerical work in terms of its applicability to the Columbus metropolitan area. In brief, the rationale behind Nelson's explanation is grounded in three suppositions:
(i) The spatial entrapment of suburban mothers with children: Due to their commitment to their domestic roles suburban women can only support short journeys-to-work (a similar argument is inferred for female clerical workers).
(ii) The married suburban mother's need for paid employ ment : Due to owning newly constructed housing and the existence of young children, many married suburban mothers need to be in paid employment
(iii) The relocation of back offices to the suburbs:
Due to the "preferred clerical labor supply" characteristic of suburban women (low-waged, well-educated, non-militant, married women with young children), firms employing large
- 139 - 140 numbers of female clerical workers (and which wish to accrue savings from employing an educated but cheap female clerical labor force) may choose to relocate to suburban residential areas of new single-family housing and of con comitantly spatially entrapped women located within easy access of highway facilities.
In this chapter the applicability of Nelson's argument to the case of Columbus will be explored by employing extensive research methods. First, an assessment of wheth er, in fact, there has been a suburbanization of female clerical labor in the Columbus metropolitan area will be undertaken. This will be done by initially examining Cen sus tract shares of female clerical workers. Then, in order to take into account the changing distribution of women of paid labor force age, location quotients (see Chapter 4 for explanation) are employed. Second, aggregate (Census tract) data are employed in regression analyses to explore the relationships between the growth of clerical work in suburban locations and measures of the need of married sub urban mothers for paid employment (measures such as home- ownership, new housing and the presence of children).74
Finally, if there are relations between the growth of female clerical workers, measures of need and suburban
7 4 The author would have preferred to have examined the relationship between the growth of female clerical work ers and the actual suburbanization of offices them selves. However, data allowing this are not available. 141 locations, do they have anything to do with spatial entrap ment and, hence, the recent arrival of offices?
The latter part of the chapter examines the journeys-to- work of clerical workers who live and/or work in a number of Columbus' suburbs in an effort to ascertain whether female clerical workers are in fact spatially entrapped. A variety of data is employed including: data on the journey- to-work of female clerical workers residing in four of
Columbus' suburbs; data on the zip-code of residence of clerical workers employed at three companies located in
Worthington or Westerville and data generated from a survey questionnaire completed by 98 clerical workers employed at one of five of the firms where the personnel manager was interviewed.
5.2 SPATIAL DISTRIBUTION OF FEMALE CLERICAL WORKERS
This section assesses whether there has, in fact, been a suburbanization of female clerical workers in the Columbus metropolitan area. Census tract data for 1950, 1960, 1970 and 1980 are employed, and these data are --necessarily-- based on the residential locations of female clerical work ers in Franklin County. This section presents and discuss es maps of the spatial distribution of female clerical workers across Census tracts for each of the Census years.
A common system of shading is employed on all the maps in order to facilitate comparison. 142
Maps 1 to 6 show the spatial distribution (in quintiles) of Census tract shares of female clerical workers living in
Franklin County. In 1950, most of Franklin County's female clerical workers live in the City of Columbus (see Map
1).7 5 The areas with the largest shares are to be found to the immediate south-east of downtown, an area about five miles to the west of downtown and another about five miles north (slightly north and east of Ohio State University).
Those areas with the smallest shares of female clerical workers are predominantly outside the city (see Map 2).
Particularly notable are those tracts which form a north- south strip along the western boundary of the County and, less clearly, a cluster in the north-east corner of the
County.
In 1960 (see Maps 3 and 4) the pattern of over represented tracts in 1950 is still quite visible, although the area to the north of the downtown has become more dis tinct and there is some general dispersal away from down town.76 The areas under-represented in 1950 are still per ceptible in the 1960 maps. For instance the western strip and north-east cluster are still visible although somewhat weaker. Map 3 shows that the downtown area is clearly
75 Map 1 approximates the extent of the city of Columbus in 1950.
76 It is possible that this is related to the growth of Ohio State University. However these data are for the residences, not the work-places, of female clerical workers. 143 FRANKLIN COUNTY SHARES OF FEMALE CLERICAL WORKERS, 1950
t o .
I
QUINTILES
0.6 - 0 .7 9 MAP 1 80URCE: U.S. CENSUS 0 - 0 .59 144 FRANKLIN COUNTY SHARES OF FEMALE CLERICAL WORKERS, 1950
see Map 1
QUINTILES
1.0 - 1.79
0.65 - 0.99 0.34 - 0.64
0 .33 80URCE% U.8. CENSUS MAP 2 FRANKLIN COUNTY SHARES OF FEMALE CLERICAL WORKERS, 1960
QUINTILES
1.09 - 3.0
0.73 - 1.08 0.50 - 0.72
0.27 - 0.49
MAP 3 0.26 SOURCE: U.S. CENSUS FRANKLIN COUNTY SHARES OF FEMALE CLERICAL WORKERS, 1960 r
m
m
QUINTILES
1.09 - 3.0
0.73 - 1.08
0.50 - 0.72
0.27 - 0.49
0 - 0 .26 MAP 4 80URCE: U.3. CEN3U8 147 becoming less well represented in terms of shares of female clerical workers, as are some more distant neighborhoods in the city.
By 1970 the distribution observed in 1950 has become less apparent. In terms of areas with higher shares, Map 5 shows that the three dominant areas in 1950 are still mani fest although they have become less cohesive. There is a concentration of high shares developing along an eastern and, more clearly, a northern wedge radiating outward from downtown. Regarding locations of under-representation, the downtown "crater" has become very pronounced. In addition to be employed as clerical workers over the study period.
However it is quite possible that the patterns in this first set of maps are, in fact, largely due to the residen tial suburbanization of women of paid labor force age. The data have therefore been re-estimated in order to hold the redistribution of women of labor force age constant. This has been done by computing location quotients (see Chapter
3 for explanation) for each Census tract. Maps 7 to 12 show location quotients across Census tracts for Census years 1950, 1960, 1970 and 1980. The maps exhibit less distinct patterns than those portraying shares of female clerical workers (Maps 1 to 6). In addition many Census tracts fall into the 0.8 to 1.19 category meaning that their share of female clerical workers is approximately the 148 FRANKLIN COUNTY SHARES OF FEMALE CLERICAL WORKERS, 1970
QUINTILES
U 0.68 - 2.0 0.51 - 0.67 0.37 - 0.50 ± 0.25 - 0.36 0 - 0 .24
80URCE: U.S. CEN8US FRANKLIN COUNTY SHARES OF FEMALE CLERICAL WORKERS, 1980
QUINTILES I 0.59 - 2.0 0.45 - 0.58 0.31 - 0.44 S 0.20 - 0.30 0-0.19 SOURCE; U.8. CEN8U8 150 same as their share of women of (paid) labor force age (a location quotient of 1.00 would mean it is exactly the same). Accordingly attention focuses on those tracts in which female clerical workers are over-represented (catego ries 1.2-1.39 and 1.4+) or under-represented (categories
<0.6 and 0.6-0.79).
Map 7 shows that for 1950 the three areas with high shares of female clerical workers still re-appear after re-estimation using location quotients. The northern clus ter (around Ohio State University) is now more pronounced, but the west and the near east clusters are less distinct.
Areas with low location quotients are again predominantly outside the city (see Map 8), but the distinct north-south strip on the western edge of the County seen in Map 2 is less apparent, and has fewer tracts in the lowest category.
In 1960, Map 9 shows that while the northern area of high location quotients is still visible (note that it has also extended slightly further north), the western and eastern clusters are weaker and much less distinct. How ever, there does appear to be a new eastern wedge of tracts with high location quotients forming further east, although it is a less cohesive grouping.
Turning to under-represented tracts it can be seen that while there is a downtown crater in 1960 it is less compact than the one observed in the 1960 share Map (see Map 3). 151 FRANKLIN COUNTY
LOCATION QUOTIENTS, 1950
■ t i l
LQ VALUE
1-2! - 1.39
08-* 1.18
016 • a7 8
Q - 0. 6 9 80URCE U.8. CEN8US MAP 7 152 FRANKLIN COUNTY
LOCATION QUOTIENTS, 1950
155476
9999999999299
783443
LQ VALUE
1 .4 - 4
1.2 « 1 3 8
0 3 - 1 . 1 0
0.6 - 0.70
MAP 8 0 - 0.69 SOURCE: U.8, CEN3US cn CD n 3 3 > z 7 s z r; o o c ■ H z . r z O O > H O C o H O) o m z H o 05 CD « « 16 illlliiaii! f
SOURCE: U.S. CENSUS 154
The tracts outside the city (Map 10) generally exhibit low location quotients, although few tracts fall into the low est category, indicating that the peripheral tracts are actually moving away from the extreme deficiency of female clerical workers portrayed in Map 8 (1950).
The 1970 Map of location quotients provides a much more vague picture than the ones presented for shares. Even the downtown "crater", so distinct in the 1970 shares Map (Map
5), is interrupted by a band of Census tracts with high location quotients.77 The northern and eastern wedges are only just visible. And, quite importantly, some of the more peripheral tracts which had low rankings on the 1970 shares Map have high location quotients (see the tract in the north-west corner of the County, for example) and a large number of tracts are classified into the middle cat egory.
A less ambiguous pattern exists for 1980 (see Map 12), although different from the one displayed by the 1980 shares Map (Map 6). First, unlike Map 6, Map 12 shows that many of the more peripheral Census tracts fall into the middle category, meaning that their share of female cleri cal workers is very similar to their share of women of
(paid) labor force age. Second, although much less exten sive than they appear on Map 6, far north and far east
77 It is possible that this is due to gentrification in the neighborhoods near the downtown area. FRANKLIN COUNTY
LOCATION QUOTIENTS, 1960
LQ VALUE
80URCE:U.S. CENSU8 FRANKLIN COUNTY 156
LOCATION QUOTIENTS, 1970
LQ VALUE
0 .8 - 0 .7 8
0 - 0.68 MAP 11 SOURCE; U.8, CEN8U8 FRANKLIN COUNTY
LOCATION QUOTIENTS, 1980
LQ VALUE
80URCE: U.8.CENSU3 MAP 12 158 clusters are discernible. Third, although distributed slightly differently and more dispersed, there is somewhat of a crater in and near the downtown area.
Overall the location quotient maps do not portray the strong trend towards the decentralization of female cleri cal workers seen in the shares maps. This strongly sug gests that the shares maps were also reflecting the subur banization of women of (paid) labor force age. This in turn suggests that the weakening central city/suburban division and the strengthening of the downtown crater seen in the shares maps were largely to do with the redistribu tion of (paid) labor force aged women. The conclusion from this exercise is that by holding Census tract shares of women of (paid) labor force age constant no strong central city/suburban difference is indicated, except, perhaps, for
1970 and 1980. This then does not strongly support the thesis of a suburbanization of female clerical workers independent of residential suburbanization in Franklin
County.
5.3 MULTIPLE REGRESSION ANALYSES
In this section multiple regression analyses on vari ables, defined across Census tracts, are employed to explore the relations between, on the one hand, the growth of clerical work; and on the other hand (1) measures of the 159 need of married suburban mothers for paid employment, and
(2) suburban locations. The observational units consist of all Franklin County's Census tracts78 for Census years
1950, 1960, 1970, 1980. Over this period the number of
Census tracts grew from a total of 97 in 1950 to 255 in
198079 reflecting the rapid growth of the County's popula tion over the study period. Variables were selected on the basis of the literature reviewed in Chapters 2 and 3 in general, and with particular reference to Nelson's (1986) study. The dependent variable is Census tract share of the total number of female clerical workers living in Franklin
County (SHRFCL).80 The independent variables are identi
78 Except tract 33, the Columbus Correctional Facility and tract 68.30, the State School for the Deaf and Blind.
79 In 1960 there were 147 Census tracts and in 1970 there were 211.
80 The precise variables Nelson uses are: percentage of white, non-Hispanic people in the population; percentage of native English speakers in the population; mean fami ly income; percentage of owner-occupied housing between 1970 and 1980; percentage of women over fifteen who are married, and who are living in husband-wife families with children; percentage of high school graduates among women over sixteen; percentage of families with incomes under the poverty level; number of divorced women (com pared to married women); and the percentage of families (with young children) headed by women. In the present study SHRCHILD and SHRWOM act in part as proxies for "percent of women over fifteen who are married, and who are living in husband-wife families with children". SHRHSL10 is used in place of owner-occupied housing built in the past ten years, as that variable is not available at the Census tract level. "Percent of high school graduates among women over sixteen" was not used as it was not available for the all Census years stud ied. 160 fied In Table 17. Note that some of these variables have also be calculated as shares (means and standard deviations are summarized in Appendix C).
Table 17
Independent Variables Employed in Regression Analyses
DIST Distance (in miles) from the center of the Central Business District (C.B.D.); I.E. from the intersection of Broad Street and High Street.
MEDINC Median Family Income (in dollars)
SHRWOM Census tract share of total number of women of paid labor force age (14 to 64 in 1950 and 1960, 16 to 64 in 1970 and 1980) living in Franklin County.
SHRHSL10 Census tract share of housing built in previous ten years in Franklin County.
SHROWN Census tract share of owner-occupied housing in Franklin County.
SHRCHILD Census tract share of total number of persons aged less than eighteen living in Franklin County.
Source: U.S. Census
The model to be estimated is:
SHRFCL = f(DIST, MEDINC, SHRWOM, SHRHSL10, SHROWN,
SHRCHILD) (1)
Substantively equation (1) postulates that tract share of female clerical workers (SHRFCL) varies with distance from the C.B.D. and with a number of socio-economic and housing variables. 161
Following from issues raised in the last section of this chapter SHRWOM is included in this analysis in order to hold the residential suburbanization of women of labor force age constant. Tract share of women of (paid) labor force age (SHRWOM) is expected to be directly related to
SHRFCL in that tracts with larger shares of women of (paid) labor force age are more likely to have larger shares of all female clerical workers.81
Distance (in miles) from the C.B.D. (DIST) is expected to be inversely related to SHRFCL in the earlier Census years, and to become increasingly less negative over time, indicating that increases in SHRFCL over time are attribu table, at least in part, to increasing propensities of sub urban women to enter clerical employment.
Median family income (MEDINC) is expected to become positively related with SHRFCL over time, especially in the latter two Census years. This reflects both Nelson's argu ment -- that the preferred labor supply consists of women who consider their wages to be a "secondary" source of fam ily income -- and the recent increased propensity of women
(especially middle-class women) to take on paid employment in order to supplement family incomes.
81 As previously noted SHRWOM along with SHRCHILD are also proxies for the percent of all women age 15 and older who live in a wife-husband household with children. 162
Tract share of housing built in the previous ten years
(SHRHSL10) and tract share of owner-occupied housing
(SHROWN) are expected to vary positively with SHRFCL. This is due to the idea that areas with newly constructed, pre dominantly owner-occupied housing, according to Nelson, are attractive locations for firms employing large numbers of female clerical workers.
Finally it is expected that tract share of persons aged less than eighteen (SHRCHILD) will be inversely related to
SHRFCL, but that this relationship will become less neg ative over the study period: this reflects points raised in the literature review regarding the increased paid employ ment of mothers with dependent children since 1950, and of mothers with children aged less than 6 since the 1970's; and the part of Nelson's explanation that relates to moth ers of young children.
Table 18 reports the Betas and the t-values resulting when SHRFCL is regressed on the independent variables,
(the zero order correlations are reported in Appendix C).
Overall the variables explain 63.0% of the variance in
SHRFCL in 1950, 64.2% of the variance in 1960, 70.4% of the variance in 1970 and 69.7% of the variance in 1980.
Several comments can be made on changes in these parame ter values over time. The first is that although SHRWOM is always the most important variable, the magnitude of its 163
Table 18
Regression Coefficients (Betas)
1950 1960 1970 1980 SHRWOM 0.824** 0.870** 0.767** 0.589** (5.275) (8.549) (14.056) (12.108)
DIST -0.077 0.006 -0.035 0.110* (-0.891) (0.099) (-0.813) (2.484)
MEDINC -0.110 -0.038 -0.092 -0.193** (-1.131) (-0.562) (-1.810) (-4.317)
SHRHSL10 -0.099 0.070 0.074 0.307** (-1.055) (1.135) (1.501) (6.773)
SHROWN 0.434* 0.249** 0.290** 0.257** (2.944) (2.364) (4.006) (4.466)
SHRCHILD -0.435* -0.367** -0.229** -0.087 (-2.674) (-3.789) (-3.343) (-1.492)
N 83 141 209 250 R SQ 0.630 0.642 0.704 0.697 F 21.606 39.984 80.196 93.144
* =significant at the 0.05 level ** significant at the 0.001 level
"effect" does decline substantially after 1960. This sug gests that the geographical distribution of women of emplo yable age is decreasinqly important in an understanding of the changing geography of female clerical employees.
On the other hand, distance from the C.B.D. (DIST) seems to be increasingly important. Recall that the sign for
DIST was expected to change from negative to positive.
Although not monotonic, the general trend ijS towards DIST increasing as SHRFCL increases and in 1980 the beta is positive and significant. So more suburban locations do 164 indeed seem to be registering increases in female clerical employment, and independently of the changing geography of women of (paid) labor force age.
Also encouraging is the fact that the distribution of female clerical employees seems to be decreasingly related to childlessness. Over the study period the Beta for
SHRCHILD becomes less negative. This indicates that the earlier situation wherein those areas with higher shares of female clerical workers were those with fewer children becomes less true over the study period. This ties in with the recent increase in the employment of mothers of depen dent children, particularly mothers of very young children, as noted in the literature review.
Turning to the housing variables, the general trend for the Betas of SHRSL10 is that they become increasingly more positive over time. This indicates that the effect of recent housing on variation in shares of female clerical workers becomes progressively more important as suggested by Nelson. Unfortunately the Beta for the home-ownership variable (SHROWN) does not perform as expected, declining in value over the time period. Likewise income levels
(MEDINC) seem to have little of the effect anticipated by
Nelson.
Generally it can be concluded that the changing pattern of female clerical workers is most clearly associated with 165 the changing distribution of women. However this relation ship does become weaker after I960, and by 1980 the other variables (with the interesting exception of SHRCHILD) also play important roles in explaining shares of female cleri cal workers.
5.4 COMMUTING IN COLUMBUS
As reviewed in Chapter 3, and noted by Nelson, tradi tional gender roles ascribe more responsibility for domes tic labor to women than men. Hence when women who are mar ried and/or mothers are also waged laborers, these multiple roles, it is argued, may serve to "trap" women in space.
Indeed women's multiple roles is one of the most frequently stated explanations as to why women work closer to home than men. So one means of evaluating the spatial entrap ment thesis is to compare women's journey-to-work with men's. Moreover it can be further hypothesized that those women with more demanding domestic roles will have shorter journeys-to-work than those women with less demanding domestic obligations. In the absence of any direct measure of the female clerical worker's domestic gender division of labor (such as hours spent on housework and child-care) domestic roles will be defined in terms of parental and marital status. To investigate this question a variety of data sets are employed in order to construct the journeys-to-work of clerical workers who live and/or work in some of Columbus' suburbs. The first data set was collected from the 1985
Polk's Suburban Directory. The sample consists of 200 ran domly selected women who live in the suburbs of Worthing ton, Westerville, Gahanna or Reynoldsburg -- where growth in the employment of women has been especially rapid -- and who identified themselves as clerical workers. The second data sets were furnished by three of the ten personnel man agers interviewed on the basis of their personnel files.
These data consist of the respective firm's clerical labor force broken down by gender and zip code of residence. In one case it was also possible to obtain the marital status of the clerical workers. The third set of data consists of the results of a survey questionnaire administered to 98 employees at five of the ten firms where the personnel man agers were interviewed (the survey questionnaire is pre sented in Appendix D).
5.4.1 Polk's Suburban Directory Data Set
Sample individuals were sub-divided into married, if the woman's husband was at the same address, and "single", if there was no sign of the women being, or ever having been, married (unfortunately this actually covers single, sepa rated, divorced and widowed women: it was not possible to 167 identify the exact marital status of these women from the
Directory). Actual journey-to-work road distances (in miles) were computed for both groups of women, and for the husbands of the married women, and are reported in Table
19. 82
Table 19
Commuting Distances of Female Clerical Workers and their Husbands.
Summary Statistics! "single" wives women husbands
mean 9.6 9.1 11.6
standard deviation 6.2 6.2 5.9
N 157 43 93
Relative Frequencies (4 mile intervals): "single" wives women husbands
0- 3.9 26% 27% 12% 4- 7.9 12 22 23 8-11.9 20 20 22 12-15.9 23 12 18 16-19.9 14 12 20 20-23.9 2 7 4 24-27.9 3 0 1
Sources Polk's Suburban Directory, Columbus, 1985
8 2 Note that the N for husbands is lower than that for wives. This is because the final sample was selected on the basis of being able to locate the exact work-place of just the women. Thus it was occasionally difficult to locate the exact work-place of the husbands when (at at a later date) their journey-to-work distances were computed. A number of conclusions may be drawn from Table 19.
First, the commuting distances for married couples indicate
that the husbands do have longer mean journey-to-work dis
tances than do the wives. This corresponds to previous
journey-to-work studies, and as reviewed in Chapter 3.
However in this case-study the mean commutes for the wives
and husbands are fairly similar, and in fact there is more
variation in the wives' commutes than the husbands'. How
ever, some limited support for the spatial entrapment the
sis jLs provided by the modal classes, in that the most fre
quently travelled distance for women, regardless of their
marital status, is less than four miles, whereas the hus
bands' modal commute is four to eight miles. Furthermore a
smaller proportion of the husbands (35%) than the wives
(38%) and the "single" women (49%) travel less than 8 miles
to work. On the other hand the mean commuting distances
for married and "single" women respectively are very simi
lar; indeed the mean commute for married women is slightly
longer than that for the "single" women. Moreover a higher
proportion of the "single" women (49%) than of the married
women (38%) live less than 8 miles from their work-places.
This is somewhat contradictory to the "conventional" view
(as well as Nelson's) that married women work close to
home. It does, however, corroborate Brooker-Gross and Mar-
raffa (1985), Madden (1981) and Hanson and Johnston's 169
(1985) findings regarding the journey-to-work and marital status reviewed in Chapter 3. On the other hand the mean commuting distances for both sets of female clerical work ers are higher than those reported in previous studies con sidering journey-to-work by both gender and occupation (see
Chapter 3).
5.4.2 Personnel Files Data
This section reports the results of a second study of the journey-to-work of female clerical workers. Three of the managers who were interviewed were able to furnish, on the basis of their personnel files, zip-codes of residence for the firm's clerical labor force. To calculate the journey-to-work distances (and therefore the firms' respec tive female clerical labor market areas) the straight line distances between the zip-code area centroids and the respective firm locations were computed. These journey-to- work distances are therefore only approximate and are almost certainly under-estimated. In this instance, how ever, where we are questioning the spatial entrapment the sis these more conservative estimates are clearly prefera ble to over -estimates.
Two of the firms were able to supply listings by gender.
One of these also provided further break-downs by marital status. This section first reports the results for all three firms and then compares and discusses those results. 170
Firm 1 is located in Worthington and employs 134 female clerical workers. Table 20 reports their estimated journey-to-work distances. The mean commute for Firm l's female clerical labor force is 7.9 miles. The the modal class is less than four miles.
Table 20
Commuting Estimates for Firm l's Female Clerical Workers
Summary Statistics:
mean 7.9 standard deviation 7.4 N 134
Relative Frequencies (4 mile intervals):
0- 3.9 48.1% 4- 7.9 20.3 8-11.9 6.0 12-15.9 12. 7 16-19.9 3.0 20-23.9 6.7 24-27.9 0.0 28-31.9 3.0 32-35.9 0.1
Source: Firm 1 personnel files
Firm 2 is also located in Worthington. It employs 620 clerical workers, 471 (76%) of whom are women. Their approximate commuting distances are summarized in Table 21.
Both the mean (13.3 miles) and the standard deviation (9.9 miles) for women are greater than the mean (11.2 miles) and standard deviation (7.9 miles) for men. The modal class for both women and men is the same: 4 to 8 miles. However 171
Table 21
Commuting Estimates for Firm 2's Clerical Workers
Summary Statistics:
Women Men mean 13.3 11.2 standard deviation 9.9 7.9 N 471 149
Relative Frequencies (4 mile intervals): Women Men 0 - 3 . 9 0.0% 0.0% 4- 7.9 33.8% 45. 7 8-11.9 10.4 12.2 12-15.9 23.1 20.2 16-19.9 9.8 8.7 20-23.9 10.8 8.1 24-27.9 4.2 1.5 28-31.9 1.5 0.7 32-35.9 3.0 2.0 36-39.9 1.5 0.0 40-43.9 1.1 0.7 44-47.9 0.0 0.0 48-52 0.8 0.0
Source: Firm 2 personnel files the frequencies clearly reflect the greater variation in the women's commutes: 12% of the women compared with 5% of the men travel more than 24 miles. Moreover, men are con centrated in lower frequency classes than women: 44.2% of women and 57.8% of men commute less than 12 miles to work.
Firm 3 is located in Westerville and employs 273 people in clerical positions, 263 (96.3%) of whom are women.83
Just under half (45%) of the female clerical employees are
83 As there were only ten male clerical workers they were excluded from the analysis. However the mean commute for all men was 11.3 miles and the standard deviation was 7.4 miles (less than for either group of women). 172 married. As only a handful of the women actually identi fied themselves as either divorced, separated or widowed, and because the personnel manager pointed out that there were more, but they prefer to be identified as single, the remaining 55% of the women clerical workers were defined as
’'single".
Table 22
Commuting Estimates for Firm 3's Female Clerical Workers
Summary Statistics:
married "single women women
mean 11.9 12.9
standard deviation 9.7 8.2
N 119 144
Relative Frequencies (4 mile ii
married "single women women
0- 3.9 20.0% 15.8% 4- 7.9 28.0 16.5 8-11.9 13.4 29.4 12-15.9 13.4 16.5 16-19.9 8.5 4.0 20-23.9 8.5 8.2 24-27.9 0.8 3.5 28-31.9 2.5 3.5 32-35.9 1.5 0.6 36-40 3.4 2.0
Source: Firm 3's personnel files. 173
Table 22 shows the journey-to-work distances for Firm
3's female clerical workers broken down by marital status.
The mean commute for married women is 11.9 miles and for
"single'* women it is 12.9 miles. The modal class for all married women is 4 to 8 miles, and for "single" women it is
8 to 12 miles. So in this case married women have a slightly shorter commute than "single" women.
Overall these results indicate longer commutes for the female clerical labor force of Firm 2 (13.3 miles) and firm
3 (11.9 for married women, 12.9 miles for "single" women) than for those women clerical workers drawn from Polk's city directory (9.6 miles for married women and 9.1 miles for "single" women). Firm l's mean commuting distance (7.9 miles) is shorter than both of the other firms and the
Polk's sample. Moreover as the journey-to-work distances for the three firms' clerical work forces are based on
"airline" distances it is almost certain that the real road distances would be greater. In addition, and as with the
Polk's sample, the results for all three firms (despite the probable underestimation) are greater than those reported
in previous journey-to-work studies considering both gender and occupation.84
84 They also indicate longer commutes for female clerical workers than male clerical workers (although the Ns for the men in Firm 3 are too small to state this categor ically) . 174
The most striking thing about these results, however, is the variation between the three firms. Women employed as clerical workers by Firm 2 commute, on average, almost
twice as far (13.3 miles) as those employed by Firm 1 (7.9 miles). Similarly, there are substantial differences in
the standard deviations of the firms ranging from 7.4 miles
for Firm 1 to 9.9 miles for Firm 3. This is also reflected
in the percent of the women commuting more than the mean
journey-to-work distances for their respective firms:
31.6% of Firm l's, 46.3% of Firm 2's and 38.6% (mar-
ried)/37.6% ("single” ) of Firm 3's female labor force.
Part of the explanation for the inter-firm variations
lies in the nature of those firms. Firm 1 is a very large
employer in Franklin County with a national and interna
tional reputation. However, its line of business is not
one that relies primarily on clerical work. This may help
explain Firm l's relatively limited clerical labor market.
On the other hand, female clerical workers at Firm 2 have
the longest commutes of all three firms. This has a great
deal to do with the fact that Firm 2 maintained at least
80% of its clerical labor force when it relocated from
downtown to Worthington (see Chapter 6 for the personnel
manager's actual comment on this).
Finally, Firm 3 is heavily reliant upon clerical work
since, 79.6% of its total labor force are so employed. Its 175 spatially extensive clerical labor market may be related to the tendency for a large number of its clerical labor force to travel from distant rural areas and small towns outside of Franklin County.
Parenthetically/ the data for the three firms also high light some of the differences between extensive and inten sive research, generalization and abstraction. On the one hand, all three of these companies share the common charac teristics of being located in the suburbs (two actually relocated there within the past six years) and employing relatively large numbers of clerical workers. On the other hand, intensive research (detailed more fully on Chapter 6) indicates that despite these similarities the firms are quite different from one another.
5.4.3 Survey Questionnaire Data
This section presents the results of a survey question naire administered to 98 employees at five of the ten firms where the personnel managers were interviewed (the survey questionnaire is presented in Appendix D). These data pro vide information on the respondent's gender,05 journey-to-
05 Although the survey was not specifically aimed exclu sively at female clerical workers, that only five of the respondents are male clearly reflects the feminized nature of clerical work. Because the N for men is so small they were not included in the analysis. However the mean journey-to-work for the five male respondents is 7.4 miles (standard deviation=7.38 miles) and 21.6 minutes (standard deviation=ll.0 minutes). The widow was also deleted. Her journey-to-work is 3 miles or 8 minutes. 176 work distance and time,86 age, family status, education, children and child-care arrangements. Table 23 provides some demographic information about the 98 survey question naire respondents.
The data were first sub-divided by marital status and then by parental status. Table 24 reports the journey-to- work summary statistics for the respondents broken down by marital status.
The results indicate that for both time and distance, married women have the longest commutes to work followed by divorced women, while single women have the shortest.
Although in this case the two journey-to-work measures yield the same rank order, it is noteworthy that the dif ferences by distance are more diverse than differences by time. The fact that married women commute longer distances than single women again coincides with the findings of
Brooker-Gross and Marraffa (1985), Madden (1981) and Hanson and Johnston (1985), but runs counter to Nelson's ideas surrounding women's journeys-to-work.
86 Because Hanson and Johnston (1985) noted that different results might be produced when travel times rather than travel distances are computed for the same data (see footnote 14) both measures were collected. The respon dents provided their journey-to-work time, and the author calculated their journey-to-work distances from information that the respondents provided in the survey questionnaire. 177
Table 23
Demographic Profile of Questionnaire Respondents.
Women Men TOTAL 93 5
AGE <25 13 1 26-30 9 1 31-35 24 0 36-40 19 3 41-45 13 0 46-50 9 0 51-55 3 0 56-60 1 0 N/A 2 0
MARITAL STATUS single 23 1 married 52 4 divorced 17 0 widowed 1 0
PARENTAL STATUS (BY AGE OF YOUNGEST CHILD) Non-parent 36 2 Parent 57 3 <6 16 3 7-12 16 0 13-17 7 0 18+ 18 0
EDUCATION (BY HIGHEST LEVEL ATTAINED) High School 26 1 Some College 51 2 College Graduate 13 2 N/A 1 0
HOUSING TENURE Own 64 4 Rent 21 0 Live with parents 8 1
EMPLOYMENT Full-time 89 5 Part-time 4 0
Source: Survey Questionnaires
Table 25 shows the respondents' commute to work in both 178
Table 24
Journeys-to-Work Distances and Times by Marital Status
DISTANCE TIME standard standard N mean deviation mean deviation
SINGLE 23 9.8 8.8 22.3 14.8 MARRIED 53 12.8 12.3 28.4 19.4 DIVORCED 16 11.0 6.7 22.6 14.3
Source: Survey Questionnaire
Table 25
Journey-to-Work Distances and Times by Parental Status
DISTANCE TIME standard standard N mean deviation mean deviation in NON-PARENTS 36 10.0 00 20.7 12.3 PARENTS <6 16 10.7 6.7 34.5 21.1 7-12 16 11.5 10.7 27.2 14. 7 13-17 7 16.3 8.4 28.6 29.5 18+ 17 11.5 7.8 25. 7 20.6
Note: Parents are sub-divided by the age of the youngest child Source: Survey Questionnaire miles and minutes by parental status. Parents are further subdivided into four categories by the age of their young est child. In terms of journey-to-work distances, women with children aged thirteen to seventeen have the longest journeys-to-work, while -- perversely -- women with no children have the shortest work-trips. For journey-to-work times women with children aged less than six have the long est work-trips and women with no children again have the shortest work-trips. 179
Unlike Table 24, using two different commuting measures produces a slightly different ordering of the parental sta tus categories. Indeed two of the categories have quite contradictory rankings: among the parents the mothers of children aged less than six work the closest to home in terms of distance (a journey-to-work mean of 10.7 miles) but take the longest time (an average of 35 minutes) to get there, while mothers of adult children have quite lengthy commutes in terms of distance (mean= 11.5 miles) but rela tively short commutes in terms of time (mean= 25.7 min utes). This difference is because most of the women with very young children include the journey-to-child-care in with their journey-to-work times.
Some of the other categories also produce results which run counter to the expectations generated by Nelson's explanation. Women with no children who, theoretically could support lengthy commutes, have the shortest trips both in terms of time and distance. While women whose youngest child is aged seven to thirteen, and who theoreti cally should have short journeys-to-work, actually have lengthy work-trips.
Retabulating the survey questionnaire responses by both marital and parental status again results in contradictory and counter-intuitive results. However many of the resul tant Ns are quite small so that these findings must be con 180 sidered with caution. They are, nevertheless, interesting and worth commenting on.
Table 26
Journey-to-Work Distances and Times
DISTANCE TIME standard standard N mean deviation mean deviation NON-PARENTS Single 20 10. 2 9.34 20.0 13.25 Married 13 9.0 6.90 20.2 10.22 Divorced 3 13.0 11.36 24.3 18.35
PARENTS Youngest child, aged under 6: Single 2 6.0 2.83 40.0 28.28 Married 10 11.2 7.29 31.2 23.64 Divorced 4 12.0 6.78 25.8 23.07
Youngest child, aged 7 to 12: Single 1 10.0 --- 20.0 --- Married 11 14.2 11.80 29.5 16.65 Divorced 4 5.9 3.52 22.5 9.57
Youngest child, aged 13; to 17: Married 4 12.5 12.50 36.3 39.45 Divorced 4 12.0 6.78 25.8 23.07
Youngest child, aged 18i or older: Married 15 15. 7 18.03 28.3 21.35 Divorced 2 8.5 9.19 15.0 18.03
Sources Survey Questionnaire
Table 26 shows that the greatest distances to work are
travelled by, in descending order, married women whose
children are aged seven or older, and divorced women with
children aged thirteen and above or with no children. The
shortest commutes in distances are associated with divorced
women with teenage children, single mothers of children 181 aged less than six, divorced women with adult children and married women with no children. The women with the longest commuting times are single mothers with very young chil dren, married women with teenagers, married women with very young children, married women with children aged seven and older and married women with adult children. Short commut ing times are travelled by widows and divorced women with adult children, divorced women with teenage children, sin gle women whose youngest child is aged seven to twelve and married and single women with no children.
The two different measures of journey-to-work length produce slightly different groupings of categories.
Although some of the categories ranking high (or low) on distance also rank high (or low) on time (such as married with no children), some categories have contradictory rank ings. For instance, single mothers work close to home in terms of distance (a journey-to-work mean of six miles) but take a long time (an average of forty minutes) to get there, and divorced mothers of teenagers have lengthy com mutes in terms of distance (mean=13.33 miles) but short commutes in terms of time (mean=18.3 minutes). Second, some of the categories with high or low rankings run count er to the expectations set up by the spatial entrapment assumption. Thus, some women with heavy domestic duties have lengthy work-trips (such as divorced women whose 182 youngest child is aged less than thirteen). Conversely
other women with less demanding domestic roles have short work-trips {for example married women with no children).
In sum these results run counter to the conventional
ideas surrounding women's journeys-to-work: women with
seemingly less demanding domestic roles (single/married/
divorced women with no children and divorced women with
adult children) have short commuting distances and/or
times. While the opposite holds for women with pre-school
aged children in terms of commuting times, and for married
women with school-aged or older children in terms of com
muting times and/or distances. In short these findings are
counter-intuitive, since neither presence of husband nor
children appear to affect the work-trip as expected.
As with the earlier findings in this chapter, these
results shed doubt on Nelson's underlying supposition of
spatial entrapment in explaining the increase in female
clerical workers in the suburbs. Moreover, and as stressed
in Chapter 4, empirical regularities are not necessarily
causally conclusive. So, for instance, the fact that a
woman commutes only a short distance to work may have lit
tle to do with her attempt to negotiate her multiple roles
(perhaps she is widowed and has adult children); whereas
that another woman is commuting a great distance might be
very much related to her multiple roles (perhaps she is the 183 sole supporter of her children and must travel some dis tance to a job which provides sufficient salary and, very importantly, adequate benefits). Accordingly the next chapter involves an interpretation of the commuting data in terms of unstructured, interactive interviews with women employed as clerical workers; and with ten personnel manag ers of firms which employ large numbers of female clerical employees
5.5 SUMMARY
This chapter examined Nelson’s explanation for the sub urbanization of female clerical workers using a variety of data sets for the Columbus metropolitan area. The first section explored whether or not there has been a suburbani zation of female clerical workers in the Columbus area by employing maps illustrating the spatial redistribution of female clerical workers. The first set of maps, employing
Census tract shares of female clerical workers indicated that there had indeed been a suburbanization of female clerical workers, thus providing some initial support for the applicability of Nelson's explanation for the case of
Franklin County. However when the maps were re-estimated by employing location quotients, in order to account for the changing distribution of women of (paid) labor force age, the patterns seen in the "shares" maps were substan 184 tially impaired. This shed some doubt on the initial con clusion that there had been a decentralization of female clerical workers in Franklin County.
Second, regression analyses of tract shares on distance and on socio-economic and housing variables, selected largely on the basis of Nelson's study, were employed.
These analyses suggested that the changing pattern of female clerical workers over time was closely associated with the changing spatial distribution of women (although this relationship does weaken somewhat after 1960). Dis tance from the C.B.D. became increasingly important, and changing it's sign from negative to positive, indicating that more suburban locations were experiencing increases in female clerical employment independently of the changing geography of women of paid labor force age. It was also found that shares of female clerical workers became decreasingly related to childlessness. This tied in with the literature on the increased paid employment of mothers with very young children. Turning to the housing vari ables, it was found that the effect of recently constructed housing did become progressively more important, as Nelson suggested. The same could not be said of home-ownership or median family income, however. Neither performed as expected. 185
The latter part of the chapter examined the journeys-to- work of clerical workers who live and/or work in a number of Columbus' suburbs in an effort to ascertain whether female clerical workers are in fact spatially entrapped.
First, data on the journey-to-work of female clerical work ers residing in four of Columbus' suburbs were utilized.
The results provided only very limited support for the spa tial entrapment element of Nelson's argument. Examining data on the zip-code of residence of clerical workers employed at three companies located in Worthington or Wes terville generally strengthened doubts regarding the uni versality of the spatial entrapment thesis. While large numbers of the clerical workers did reside fairly close to their work-places, there were also large numbers of people traveling from all over the city, and from distant areas outside of Franklin County.
Finally the results of a survey questionnaire were pre sented. These largely confirm the over-generalized nature of the spatial entrapment thesis. Besides undertaking lengthy commutes, the categories of respondents convention ally viewed as the most entrapped (for instance, mothers with young children) actually had long journeys-to-work.
On the other hand, women who, conventionally, should have been less spatially entrapped (non-mothers and women with adult children, for instance) were found to have short journeys-to-work. This suggests that, at least for the
Columbus area, the evidence is not especially strong with regard to women being trapped in their place of residence, waiting for employment opportunities to come their way. CHAPTER VI
WOMEN'S MULTIPLE ROLES AND SPACE: A QUALITATIVE
ANALYSIS
6.1 INTRODUCTION
This chapter consists of an interpretation of the rela tionships revealed by the Census, journey-to-work, and questionnaire data presented in Chapter 5. This is achieved through the use of the unstructured, interactive interviews which were conducted with twenty-eight female clerical workers living in Westerville and Worthington; and with ten personnel managers of Westerville or Worthington firms which employ a large number of clerical workers (see
Chapter 4 for a full discussion of the methodology and rationale behind the use of interviews).
Following the introduction, the second section describes the interview data employed in this chapter. The third section summarizes the spatial entrapment thesis and pro vides illustrations from the interviews to highlight the spuriousness of linking women's short commutes with heavy domestic reponsibilities. The remainder of the chapter outlines an alternative explanation of the women's journey-
- 187 - 188 to-work patterns to that offered by the spatial entrapment thesis. Drawing on the conceptualizations presented in
Chapters 2 and 3, this alternative explanation partly con sists of an examination of the changing powers and needs of women in paid employment in recent decades. The women interviewed have the distinct ability to plan and learn new ways of acting and so possibly transform both their powers and needs, and their socio-spatial system. Thus a central concern in this chapter is an investigation of how the interviewees have created coping strategies and even altered their powers and needs so as to better negotiate their multiple roles.
While the women interviewed actively alter their socio- spatial system, their ability to do so is clearly mediated by wider changes in society. Accordingly attention also concentrates on the changes in the infrastructure of insti tutions and practices which have enabled more women to bet ter negotiate their multiple roles. This is followed by a discussion of how to explain women's journeys-to-work.
Attention focusses on how the women's commutes reflect their changing powers and needs, the spatial patterns of job opportunities, and the extent to which they are enmeshed in an existing set of localized relations. The results of interviews with the personnel managers of ten firms are also presented, paying special attention to the 189
spatial aspects of their labor markets. The chapter closes
with a summary.
6.2 INTERVIEW DATA
Twenty-eight women were interviewed and Table 27 presents a brief profile of them.
Table 27
Profile of the Interviewees
EMPLOYMENT NAME AGE MARITAL STATUS CHILDREN STATUS
Alison 23 single (co-hab) none full Paula 26 single none full Christine 26 married none full Andrea 27 married 2(5/3) full Clare 31 married 1(2) full Tracy 34 married 2(12,10) full Yvonne 36 married 2(8, 5) full Lynne 37 married 2(16,12) part Kathy 38 married 2(13,16) full Jamie 38 married (2nd) 4(7,11,14,19) part Sally 41 married (2nd) 2(19, 21) full Wendy 42 married (2nd) 2(6,13) full Penny 44 married 3(23,23,18) full Lesley 40 married 2(18,17) full Liz 41 married (2nd) 2(7, 15) full Ruth 45 married 3(21,19,12) full Audrey 46 divorced 2(20,18) full Marion 47 married 4(18,20,24) share Sheila 48 divorced 3(18,15,12) full Val 51 married 4(29,27,24) full Rose 53 divorced 4(19) full Jean 55 married (2nd) 6(19,12) full Sylvia 58 married 2 full Marcia 58 married 3 full Wanda 59 married 3 full Olive 61 married 4 part Meryl 62 married none temp Marge 64 married 3 part 190
They were selected from the sample of two hundred women
drawn from the Polk’s Suburban Directory data referred to
in the previous chapter. As Worthington and Westerville
have been sites of recent and major office location, only
those women who live (but do not necessarily work) in
either of these suburbs were considered. Actual selection
was based on obtaining a varied sample according to the
following criteria: marital status, motherhood, and the
location of and size of the labor force of their employer.
Materials introducing the researcher, explaining the
research and the importance of the interviews, a list of
possible interview items and a sample interview, were
mailed to the women. The women were later contacted in
order to arrange an appointment to interview them. The
majority of the interviews took place in the women's homes
-(three took place at the women's work-places), were tape-
recorded unless the women requested otherwise, and lasted
approximately three to five hours (see Appendix A).
Interviews were also conducted with the personnel manag
ers of ten firms which employ large numbers of female cler
ical workers, and are located in either Westerville or
Worthington. The companies were largely selected from the
Columbus Area Chamber of Commerce Membership Directory on
the basis of the number of clerical workers they employ and
the type of business they were engaged in (see Appendix B). 191
These interviews provide a greater understanding of the particular types, of necessary relationships or causal structures which exist and bear upon the research question.
Furthermore they identify the particular conditions of the interviewees and the contingent circumstances of their lives, which give empirical expression to those necessary ways of acting. A number of important issues about the paid employment of women are explored, namely: (i) how do women solve the problems of their multiple roles (e.g. commuting time, part-time work, and how this is related to the presence of children, marital status, etc.)?; (ii).the nature of the forces pushing women into, and facilitating entry into, multiple roles (e.g. marital status, attitudes to working women, availability of child care facilities);
(iii) whether, and the degree to which, there have been changes in the gender relations, roles and ideology promot ing the entry of women into paid employment; and (iv) what role (if any) the employer plays in these issues.87
87 The personnel managers were also asked questions related to spatial structure, such as the reasons why their respective firm located where they did, and the charac ter of their clerical labor market areas. 192
6.3 THE SPATIAL ENTRAPMENT THESIS
Numerous authors have documented a trend towards the suburbanization of female clerical work. It is suggested that this is related to the relocation of firms employing large numbers of clerical workers to suburban destinations to take advantage of low-wage, well-educated pool of female labor. This explanation for the changing geography of female clerical employment is based on the premise that suburban women, especially those who are married and have young children, are captive in their place of residence.
It is argued that because of their commitment to their domestic roles as caretakers of the home and children, these women are willing to forego well-paid jobs in favor of locally available jobs which allow them to attend to their domestic obligations. Thus one would expect that the greater a woman's domestic responsibilities, the shorter that woman's work-trip would be.
However the journey-to-work studies in the last chapter did not bear this out. Indeed some of the women with heav ier domestic obligations had among the lengthiest commutes, while the opposite held for some of the women with few domestic obligations. Using illustrations from the inter views with women clerical workers it will be argued in this chapter that the multiple roles of women imply no necessary relationship to space. 193
A number of the women interviewed do have short
journeys-to-work. However in many cases the women's short commute is completely spurious in that they are not the
spatial outcomes of the women's efforts to combine their
various roles into a finite period of time. For instance,
negotiating multiple roles does not really play a part in
the resultant short work-trip of Andrea, whose children are
three and five. Although she was pleased to be offered a
job only five miles from home, this was not due to having
to be close to home to deal with her children as they are
taken care of by her partner (who works at home) and her
father. Instead it was because she would be working with
her mother and, if she decided to, she could go home for
lunch and spend some time with her partner and children.
Gail starts school this year, she was in pre school two days a week. My dad's been retired for ten years now and he watches them the rest of the time. Plus for the last two years my hus band's office has been in our basement, so he's actually here most of the time too. Anyway, I drop her off at pre-school, and Joey's with his dad first thing, then my dad comes and gets him. It's been so helpful the way my husband's work is set up and with my dad. I don't have to take time off if the kids are sick, because Mark (her part ner) or my dad are around (Andrea).
In other words, Andrea's domestic arrangements are such
that the children are largely the responsibility of their
father and grandfather (at least on week-days).
In other cases a short commute is also spurious because
it is no longer an attempt to negotiate the roles of paid 194 worker, mother and partner. At an earlier point in the
women's lives it may have been, but because of, for exam
ple, the exit of children, it has lost the meaning it once
had. This is the case for Olive, Marcia, Wanda, Marge and
Audrey, whose commutes are all less than two miles, but who
no longer have dependent children living at home.
Some of the women interviewed had both young children
and a long journey-to-work which is entirely contradictory
to expectations generated by the notion that mothers with
dependent children are spatially entrapped. For instance,
Liz, whose children are now fifteen and seven, only took a
few weeks off for their births. For the last fifteen years
she has worked in or near the downtown area. Until her
second marriage eight years ago (she was divorced in 1976
and got custody of her son, who was three at the time) she
lived in Bexley with a five mile commute to work. When she
and her second husband moved to Westerville, she kept the
same job and increased her work-trip to almost nineteen
miles.
I wouldn't take a job in Westerville just because it's near home, that has no effect on me. When I married my second husband and we moved to Wester ville, I stayed with this job because of the job itself. I'd never change jobs just because of location, unless I was driving for over an hour. Because this job pays what I want and it's men tally stimulating I haven't changed and won't change because of moving house (Liz). 195
Then there are the women whose journey-to-work has actu ally become shorter as their children grew older. For instance, when Sally first started her job ten years ago, her sons were nine and eleven and she used to commute eighteen miles. Then she remarried four years ago and moved to her partner's house in Westerville which decreased her journey-to-work to four miles (her partner worked downtown and it would be too far for him to commute from Delaware, where she used to live). In other words, as with most of the interviewees, Sally's residential locations vis-a-vis her work-place have been affected by a whole variety of considerations other than just her multiple roles.
Finally consider the case of Kathy, with children who are now sixteen and thirteen, and who has worked for the same firm for thirteen years. For the first year she and her partner lived about three miles from where she worked.
Then they bought their house in Westerville which did increase the length of her work-trip, but given her child care arrangements the move actually attenuated her multiple roles, something which would simply not be apparent by looking at her change of address:
When I first started my job thirteen years ago we lived about ten minutes away (from where she works). But it took me an hour because I had to take the kids to the sitter and she lived (about ten miles away). So I was getting up very early because I began work at seven. After a year we moved to Westerville, so the sitter was actually on my way then, that made things much easier (Kathy). 196
It is clear from these interviews that a short journey- to-work is not always tied to a woman's negotiation of her multiple roles, nor is a long journey-to-work necessarily characterized by the absence of such a negotiation. In brief, the journeys-to-work of the women interviewed estab lish that, regardless of their marital status and the pres ence of children, they are not particularly spatially entrapped.
6.4 AN ALTERNATIVE EXPLANATION OF WOMEN'S COMMUTING
PATTERNS
Given that the spatial entrapment thesis is an unsatis factory explanation of the journeys-to-work of the women featured in the journey-to-work studies in the last chapter and the interviews being employed in this chapter, how can the women's commuting patterns be explained? The remainder of this chapter offers an explanation of their journeys-to- work alternative to that offered by the spatial entrapment thesis. Drawing on the conceptualizations presented in
Chapters 2 and 3, this alternative explanation partly con sists of an examination of the changing powers and needs of women in paid employment in recent decades. These changing powers and needs can either facilitate or impede women's attempts to negotiate their multiple roles. At the same time it must be recognized that the interviewees have the 197 distinct ability to plan and learn new ways of acting and so possibly transform both their powers and needs, and their socio-spatial system. Hence the main thrust of this section will be an investigation of how the interviewees have created coping strategies and even altered their pow ers and needs so as to better negotiate their multiple roles.
6.4.1 Divorced Women
As detailed in Chapter 2, divorce can impose new needs on women. Some of the divorced women (Sheila and Rose) and women who have been divorced but have since re-married
(Sally and Wendy), remarked that their divorces were the initial reason that they went back to work (they had had paid jobs before they were married), or began to work full time rather than on a part-time or temporary basis. As
Sally said:
I was put in the position of getting a divorce and so I went to see my boss and said this can't be on a temporary basis anymore. I needed insur ance for myself and the children. I needed ben efits, and life insurance, I needed to put in for a retirement fund. It hadn't been a concern before. So they gave me a full-time job (Sally).
Theoretically divorce could also present women with new possibilities, such as being able to relocate closer to the work-place (as a partner's needs no longer have to be con sidered in residential location decisions). However given the openness of the women's socio-spatial system, other 198 factors must be taken into consideration. For instance rather than move to a neighborhood closer to work, Sheila commutes the ten miles she presently drives in order to both secure an adequate salary and benefits to support her children (aged eighteen, fifteen and twelve), and because having been awarded the family home she says that:
I'm determined to hold on to the house, If we moved it would disrupt the children's lives, they'd have to change schools, they'd lose their friends, I couldn't do that to them, I'd hate myself. So many divorced women end up having to sell the family home and move into apartments, that's not going to happen to me. I'm proud of this house and the marriage it used to represent. I don't want my children to suffer because of the divorce, so the house is really important (Shei la) .
To some extent it appears as though the increasing rate of divorce in recent decades has also played a part in intensifying women's career aspirations, especially among younger women. The needs created by divorce appear to have prompted some of the interviewees (mostly the younger women) to plan for the possibility of divorce. Thus although many of the younger interviewees (married or not) said that they would like to be married for life, they realize that this may not be the case. Moreover they do not assume that their partners (or prospective partners) will
"keep them". Partly this is because they are aware that the economic impacts of divorce are more severe for women than for men, and partly it is because they simply do not wish to be financially dependent upon their partners. 199
Divorce is something to worry about, because now you cannot plan on anything. You have to worry about yourself and your children if you have them. Just because you're married is no guaran tee that its going to stay that way. I can count on one hand the number of women at work who are still married to their first husband, the rest are divorced and remarried. Many of them are raising children on their own, which is their main purpose for being there. These women aged thirty-five or more had this romantic idea that they didn't need a career, they'd just marry their high-school sweetheart and they'd look after them, they'd have a family. Then the bot tom falls out of their world, and there they are with few job skills, with children, and you know few fathers pay child-support, and anyway that's not enough to raise a child (Sally).
My husband just told me we can afford for me to quit work or work part-time if I want, but I don't want to. My friends say I'm cynical, but I think I'm being practical, what if he dies or runs off with some bimbo (Christine).
Jean provides an interesting variation on the relation ship between divorce and career aspirations. Her second husband's alimony and child support payments to his former wife put such a strain on their family income that Jean's promotion and concomitant salary increases became crucial to their budget.
6.4.2 Domestic arrangements
In theory one coping strategy to ease the strain of mul tiple roles (at least for the women who live with their partners) is the degree to which they could enlist the aid of partners and children in the domestic division of labor.
In practice, and as studies reported in Chapter 2 illus 200 trated, this may be difficult to achieve, especially among older couples. So it was no surprise to find that some of the interviewees' respective partners, usually among the older women, did very little in the way of domestic duties.
However some of the interviewees, usually among the younger women, felt that their partners did a great deal, especial ly compared with their fathers. As expected the most equ itably divided task was that of child-care, particularly among the younger couples. However when it came to the more mundane tasks of housework, most of the women, regard less of their age, felt that they did more then their "fair share".
My husband does not help in the house. He does better at things like taking the kids to the park. He's often out there with all the neigh borhood kids. If I asked him to hang a picture or wash the floor he'd look at me like I'd asked him to go to the moon. He never learnt that stuff, he isn't interested, and to be honest I don't think-he'd do it to suit me (Sylvia).
Some of the women did stress the importance of their older children in the domestic division of labor. They talked of how they had got their children to "help" around the house. For instance Marion's children (who are in their teens and twenties) have been assigned household tasks for at least the last eight years. Marion expects these tasks to be completed before they go to bed. She says her philosophy is: "You eat, therefore you cook and wash-up; you wear clothes, therefore you wash them and sew 201 on buttons". It appears that older children are especially important to divorced mothers. Two of the divorced women said their teenage children had taken part-time jobs in part to reduce the economic burdens they placed on their mothers. One of Sally's sons regularly pays the utility bills, and one of Sheila's sons often pays for the grocer ies .
Many of the married women have made attempts to alter their domestic roles. Child-care, as noted earlier, seems to have become more equitably divided among couples, but there are many other coping strategies including using a diaper service and eating out at restaurants; as Christine commented: "When we can't be bothered to cook, we just go through the drive-through at Taco Bell". And Sally,
Yvonne, and Sylvia have hired domestic help. Most of the women talked of lowering their own expectations in that they became less fanatical about having a "Metropolitan
Home" house, like Lesley who remarked: "When I see dust now, I just leave it, other things are more important now".
Kathy even renegotiated a new, more equitable division of domestic labor between herself and her partner.
I was doing everything, I thought that I should be "Supermom" I didn't realize I had the right to ask my husband it's time for you to take the kids out, do the dishes and so on. Then five years ago I decided I couldn't handle it any more. The kids always called me, and I was the one taking time off for their doctor's appointment. As I've moved up the ladder, I can't just take time off. I was the one doing all the cleaning while they 202
were out, if I was lucky I had a couple of hours on Sunday evenings. Now he does more than half. We got counselling, because I said if this is what marriage is about then I've had enough. We decided we didn't want to break up the marriage, so we got counseling and got it all ironed out. He's stuck to his bargain and I've stuck to mine. Whoever gets home first, starts dinner, who ever is in the basement puts a load of laundry in. It's no longer mom's job (Kathy).
However, when it came to the question of whether domes tic (and for that matter, reproductive) technologies have make it easier for women to fulfil their domestic duties and go out to work, it was more likely to be the women in their sixties, like Meryl, who agreed. The younger women had more taken-for-granted responses when questioned about the part played by domestic and reproductive technology in their being able to add the role of waged laborer to their domestic roles. Presumably, these changes have had a greater impact on older women simply because they remember what it was like before they had automatic washers and dri ers, microwave ovens and reliable birth-control, as Meryl clearly illustrates:
If I could have a dime for every one of his shirts that I've ironed over the years ... and the young girls in the offices today tell me that they never iron anything. They don't realize how wonderful drip-dry clothes and tumble-driers are... when I was young, when you got married that was it, we knew the babies would come soon after, we didn't have the Pill to take back then. So once the babies came you had a full-time job in the home (Meryl). 203
6.4.3 Career aspirations
It is possible that one of the reasons that women, espe cially married women with young children, have convention ally been viewed as "trapped" in space is that gender ideology has helped segregate women into low paid, low sta tus jobs, which do not pay sufficiently well enough to
"justify" lengthy commutes. However this seems to be becoming less appropriate with the increase of women moving into all fields of paid employment, the rise of the "career woman", the ongoing (political, social and legal) struggle to ensure equality for women and changing attitudes towards women in paid employment. Indeed as pointed out in Chapter
2 the wages of married women have become increasingly vital to the maintenance of a middle-class standard of living for their families.
However, while most of the women interviewed did say they were working for economic reasons (their salaries are vital to their family incomes), it was clear that many of the women interviewed also consider their jobs to be more than just a paycheck. Many see themselves as career women, deriving a great deal of self-identity and self-confidence from their paid employment. This helps to explain why some of them have chosen seemingly complicated, expensive and even (to the naive observer) illogical solutions to the problems presented by multiple roles. For instance, Rose 204
faces a twelve mile commute to work, where her wages as a clerk-typist do not seem high enough to compensate for her
long journey-to-work. However, she rationalizes this in
terms of the economic stability of her employer, and the
extensive opportunities for promotion that she anticipates
exploiting in the future.88 Paula feels the fifteen mile
commute that her new job-entails (in her previous job she
only drove six miles to work) is only inconvenient in that
she has to get up fifteen minutes earlier than she used to,
which she feels is acceptable given the greater prestige,
higher salary and better career prospects associated with
her new job.
Indeed being career oriented was particularly prevalent
among the younger women, like Paula and Alison. They talked
of being brought up to expect to earn their living and to
pursue a career. They did not envisage marriage or mother
hood as a signal to retreat from the paid labor force.
They knew it would be difficult at times, perhaps they
would switch to part-time employment for a few years, but
they certainly saw themselves working while raising their
children. This is precisely what Clare, Yvonne and Andrea,
mothers of pre-schoolers, have done. All three took a min
imum amount of time off after the births of their children
and, except possibly switching to part-time for a short
88 Interestingly, since my interview with Rose she has in fact been promoted numerous times and now holds a well paid position with a great deal of responsibility. 205 time, do not imagine that they will give up their careers.
Among the group of older women who described themselves as career women it was apparent that their jobs had not always been careers. For some their self-image had gradu ally changed as, for instance, their children grew older and less needy of their mothers, as it did for Jean, Penny and Lesley. In other cases it has been related to divorce, as it was for Audrey and Rose, or a lack of marriage pros pects as in Meryl's case, who did not marry until her mid thirties .
I do think of myself as a career women, but it hasn't always been that way. It changes because your life changes. I had always been a wife and mother, but now my youngest will be leaving next fall for college and my husband travels. It's just changed. A lot of the time I used to spend going to ball games and the time I used to spend with the kids is extra time now. I enjoy working, I enjoy the people contact, I don't enjoy staying home, cooking and cleaning. I've done all that and I did enjoy it. I worked hard during those years I had them all at home, but I don't want that anymore (Penny).
When I got married I simply never intended to go out to work again. But thirteen years later I found myself divorced, with four kids to raise and I no longer had my husband's pension to look forward to, so I had to work. Originally it was for money for me and the children. Later as the children have gotten older I've become more interested in getting a good job, with promotion prospects and a good pension plan. I'm working for me now (Rose). 6,4.4 Work Schedules
Historically a "classic" time-space budgeting strategy for married women, especially those with school aged chil dren has been part-time or temporary work. Yet, regardless of the presence of children, most of the women interviewed worked full-time. The reasons they usually gave for not working part-time (although some have done so in the past) included the limited pay and benefits associated with part- time work, lack of promotion prospects, and personal pref erence for full-time work: as Tracy said "I need full-time employment to satisfy me". But it was also the simple lack of appropriate part-time work. Despite the prevalent assertion in the literature that the increase in the avail ability of part-time work is related to the increase in women's paid employment, only six of the women interviewed did not work full-time (they either worked part-time, job shared, or worked through temporary employment agencies).
And of these six women, three of them have children living at home (Lynne, Marion, and Jamie) but most of these chil dren are teenagers or older; the other three, Olive, Marge, and Meryl, either never had children or their children have already left home. These three women work part-time in order to save some additional money for their retirement
(Olive and Marge's partners have already retired and
Meryl's will do so in the next year). 207
However, while few of the women worked part-time at the
time that they were interviewed, some of them had worked
either part-time or as temporaries when they first went
back to work or while their children were growing up. Take
Val89 and Sally for examples
I wanted to work, but not full-time, I did the part-time because I was still needed here at home, my oldest was a senior in high school and I started to work in case they wanted to go to col lege, which they all did. I noticed that most of the part-timers they hired were in my age group. Maybe it's because when we go to work we give them a day's work, better than youngsters in their teens and twenties. Yes, most of us part- timers were in our late thirties and forties, the full-time people were any age. Most of the part- time people were women too, every now and again there would be a man, but he was usually a col lege student (Val).
When I first started back and the boys were still small, I temped. It was great you could work in the winter, working when you wanted. Then when summer comes around and the kids were out of school you can say I'll call you in September. Temping is really a great way for a woman who hasn't worked in years to get her feet wet, you learn about them and they learn about you, and sometimes you get hired permanently like I was (Sally).
89 Val's employer happens to be one of the firms where the personnel manager was interviewed. This firm does have a policy of hiring older women to work part-time pre cisely because they believe them to be more reliable, more productive, and less prone to absenteeism and turn over than other groups of people who work part-time, such as college students. 208
6.5 SOCIETAL CHANGES
While the women interviewed actively alter their socio- spatial system, their ability to do so is clearly mediated by wider changes in society. Thus changes in the infra structure of institutions and practices have enabled more women to better negotiate their multiple roles. Two impor tant changes which have taken place in the recent past are the increase in child-care facilities and the apparent increased flexibility of employers towards their female employees. These changes are considered in the next two sub-sections.
6.5.1 Child-care Facilities
The women with young children like Clare, Andrea, and
Yvonne, were particularly concerned about the availability of affordable, quality child-care. Unlike most of the women aged forty-five and older, who stayed home while their children were growing up, the younger women have tak en, or plan to take, very little time off work for child rearing. As Clare said:
I think the most important thing firms could do for us is to provide us with on-site child-care. I think women would be so much more productive if they knew their child was being well looked after and was near by so they could visit on their breaks. I don't get to see Kyle during the day, but he's at an accredited child-care center, and though I still worry about him I can see the dif ference his being there is making. He's very confident and outgoing, it's doing him good (Clare). 209
Moreover the number of child-care centers in communities have expanded rapidly since the sixties. Hence women such as Kathy and Liz, who have worked since their children were young, commented on the difficulties that they used to face in their efforts to secure good and reliable sitters.
We had to get a private sitter, I knew of this woman from my old room-mate from my single days, who took her child to her. They went to the sit ter until they were twelve. I took them, she used to live (ten miles away), but she was good and I just didn't know anyone else who would take the kids. But it was worth it because she was so good, a friend recommended her and you don't like to leave your baby with a perfect stranger. We didn't have these nurseries like you do now, I think they're terrific, they're in spots close to where you work, they're great (Kathy).
Indeed given that women with children aged less than six have become the fastest growing sub-group in the paid labor force, and that the ten firms interviewed did employ many such women, it was suprising that none of the firms offered on-site child-care. Although in the last few years two of the firms have begun to offer their employees vouchers for nearby child-care centers and a child-care option in their benefits packages packages. However most of the personnel managers did say that their firms had either recently con sidered or were presently considering, provisions for such a service (although many commented that liability insurance might make it prohibitive)!
We have one benefit that we call a flexible ben efit, which you get on a pre-tax basis. You can set aside extra money before the government taxes it to cover things which aren't covered and you 210
can put in as much or as little as you want, within certain rules. Some people put in for their child-care this way, and others take the vouchers we offer for the child-care center here in (the industrial park).
We're not very progressive, I wish we were. I really respect the firms who offer some help in the day-care. We're really behind the times in not offering something along these lines, subsi dizing it, or offering actual facilities, some compensation of some sort, even time off or more flexibility in the scheduling. There's a day care center right next door. Last year I even tried to set up an arrangement with them for our employees. Unfortunately the road-block I'm fac ing is that unless I can offer a benefit for everyone that will benefit the majority, we can't have it. But it's brought up time and time again, "When are we going to have day-care or something", so I think we'll have it some day soon.
6.5.2 Employer Flexibility
It appears that in the last few years the ten firms where the personnel managers were interviewed have become much more flexible about their female employees taking time off to attend to family or personal needs. Indeed most of the firms noted that the women who were mothers, especially the ones who were divorced, were among their better employ ees in terms of having lower rates of turn-over and absen teeism. In most cases arrangements were made with depart ment supervisors and, as long as it did not become a chronic problem, the women did not usually face any resis tance from their managers. As one personnel manager noted:
We try to work it out on an individual basis with our supervisors. We recognize those (family) needs, but we also want acceptable attendance. 211
We allow people to make up the time, and we ask them to schedule things early in the morning or late in the afternoon, to minimize the time they are absent. They make up their time or we just forget about it. But that's on an individual basis worked out with each supervisor.
Given an apparent willingness to accommodate their female employees, and the impression that part-time work, at least for married women with children, is the "classic" coping strategy, it was also suprising that few of the firms offered the possibility of part-time work, and even among those that did there were a limited number of posi tions. Generally it was felt that there were very few jobs which could be completed in part-time hours; and what part- time work there was was heavily biased toward entry level positions which the personnel managers feel do not attract too many people.
6.6 DISCUSSION
The spatial entrapment thesis, upon which much of the suburbanization of female clerical work literature is based, assumes that suburban women, especially those who are married and have children, are captive in their area of residence. It is argued that because of their commitment to their domestic roles as caretakers of the home and chil dren, these women are willing to forego well-paid jobs in favor of locally available jobs which allow them to attend to their domestic obligations. Thus one would expect that 212 the greater a woman’s domestic responsibilities, the short er that woman's work-trip would be.
However the various empirical studies in this disserta tion do not support his. Indeed some of the women with heavier domestic obligations had among the lengthiest com mutes, while the opposite held for some of the women with few domestic obligations. In short it would seem that the multiple roles of women imply no necessary relationship to space. This is because the socio-spatial system in which a woman in the paid labor force operates is not a closed one.
Her work-place location is affected by more than just her place of residence and vice versa. For instance, other important spatial factors include the locations of her child-care arrangements and/or her children's schools, school districts, and (if she is married) her partner's work-place. These may be manipulated so as to negotiate her multiple roles but a short journey-to-work is not nec essarily the only possible spatial outcome of this. Indeed women's journeys-to-work should be conceptualized as part of a much larger time-budgeting problem in which the women can make adjustments to better suit their needs. Some examples of this from the interviews include child-care arrangements where the woman's partner or another relative takes care of the children and child-care arrangements (as with Andrea and Liz, whose partners attend to the children, 213 at least on week-days; and Ruth whose two older children schedule work and school around their younger brother's « school day); doing grocery shopping and bulk cooking (and freezing) of meals for the following week at the week-end
(as Rose/ Audrey and Marge do); and hiring household help
(as Sally, Yvonne, and Sylvia have done).
The basis of the spatial entrapment thesis is the con cept of an empirical regularity which relates women's gen der roles to spatial limitations, particularly in terms of women's commuting distances and job search areas. As explained in Chapter 3, empirical regularities are not especially useful in explanations of human activities, giv en the indeterminate relation between social structures
(and hence gender roles) and spatial patterns. In other words, the spatial expression of particular, socially embedded powers and needs is contingent upon pre-existing spatial patterns. So, for example, where job opportunities are limited people will travel further to reach areas with better opportunities.90 Hence in understanding the actual commuting patterns reported in this dissertation (which appear to invalidate the spatial entrapment thesis) one has to consider not only the changing powers and needs of women but also the spatial patterns of job opportunities.91
9 0 This has been recognised in mainstream human geography in the debate over gravity coefficients, but it was also recognised, more implicitly, in earlier work on inter vening opportunities. 214
At the same time one might ask why don't the women move closer to their work-places? Or alternatively, when the firm relocates to a less accessible location why do many of the clerical employees keep their job and extend their journey-to-work rather than finding a substitute closer to home? The answers to these questions appear to stem from how the women are locked into an existing set of localized relations. For instance, they may find themselves tied into good child-care arrangements and/or the local school system, they may be limited by the location of their part ner's work-place, or by low mortgage payments.
Conversely the job itself may be a source of relations which are difficult for the women to find substitutes for.
Many of the women who considered their jobs to be at least as important as their domestic roles, talked of being reluctant to change jobs because their relatively high sal aries and promotion prospects were linked to a well- established employment history with their current employer.
Indeed the internal labor markets of their employer may, in effect, bind them to that particular firm. All ten of the firms interviewed do have some sort of internal promotion system for their clerical employees. So, for instance, in the search to fill a position, present employees are con sidered before the firm takes the search to the external
91 This is especially apparent in the long commutes of the women coming from rural areas to work at the three firms which provided zip-code listings (see section 5.4.2). 215 labor market.92 In addition salaries, benefits, vacation and sick leave are all tied to both seniority and length of service. As one personnel manager explained:
We really believe in rewarding people who choose to make a career with us. So for example, our profit-sharing is based on the number of years of service with the firm. Pension, pay, vacations, sick pay, they all depend on years of service. While other things, like life insurance, are tied to your salary it goes up as your salary goes up, so there's some incentive to advance yourself.
It was also noted that a positive outcome of the creation of an internal labor market for most of the firms has been a decline in the turnover and an increase in the productiv ity of their clerical workers. Related are many of the firms concomitant attempts to "boost morale" through such things as firm outings for the employees and their fami lies, and what one personnel manager referred to as "recog nition" .
We started recognizing people for making a career with the firm. In my department I make a big deal about their anniversaries (of when they began working for the firm), I take them out to lunch and make a fuss about changing their plaques (each person has a plaque with their name and length of service on their desk). It really means something to us, you can recognize people for doing a good job, for going another year with the firm. You can give them praise, pat them on the back.
92 In all but two cases the firms interviewed had intro duced internal career ladders within the last ten years. Two personnel managers pointed out that the recent nature of this firm policy is directly related to chang es in federal and state legislation regarding the Equal Employment Opportunities Commission (EEOC). 216
So it appears, according to both the interviews with the women and the firms, that internal career ladders, along with women's career aspirations, tend to bind the women to the firms. Indeed, this prevails even when the women's various roles change in such a way as to indicate that switching jobs might be an appropriate response. Thus, women have maintained their jobs despite becoming first time mothers, children leaving home, remarriage, residen tial relocations which lengthened work-trips, and even when the firm itself relocated to a site less accessible to their homes.
A possible implication of this analysis, however, is that there are no limits to the stretchability of spatial relations: if a woman's particular job is as necessary as her residential location, then even when the distance between the two increases (due, for example, to the job relocation of her partner or the relocation of her place of employment) she will still maintain the connection and sim ply elongate her journey-to-work. Clearly this is not the case. There are limits to spatial stretchability. Thus, how can these dilemmas be resolved?
Presumably, depending on the individual and her/his cir cumstances, some relations are more necessary than others.
For instance, a single mother's job is relatively more nec essary than that of a married mother. Alternatively a job 217 which was initially contingent for a woman may acquire a degree of necessity over time. So, for instance, some women continue to work close to home even though they no longer "need" to (for instance their children have grown up or they get divorced), while others have long journeys-to- work even though (theoretically) they should only be able to support short commutes (for instance, those women who fit the "preferred clerical labor supply" stereotype of being married with young children).
Moreover, it is through real social conflicts that these different degrees of necessity are negotiated: is it more necessary to a woman that her partner be happier in his job than she is in hers? Is it more necessary that a woman's relatives feel good about her rather than that she has waged employment? It is through conflicts such as these that social structures and spatial patterns are changed: which returns us to the role of changing powers and needs in understanding the changing character of women’s journey- to-work.
6.7 OFFICE LOCATION AND SPATIAL ENTRAPMENT
As previously noted, the prevailing view in the litera ture dealing with the changing geography of female clerical work, is that firms relocate to the suburbs in order to take advantage of captive pools of low-wage, well-educated 218 female labor with heavy domestic obligations which forfend lengthy commutes. Three of the ten personnel managers agreed that this had been one of the more important reasons for relocating to either Westerville or Worthington. How ever this was not an important aspect of the relocation decision of the other seven. Indeed this suggestion pro voked some surprising comments:
Well, that could have been true maybe ten or fif teen years ago, but not now.
We moved about fifteen or sixteen miles (from downtown to Worthington), but I'd say about eighty percent (of the clerical labor force) stayed with us. We hired the extra twenty per cent from around here. The demographics of who we lost were largely those people who hadn't been with us for that long, less than five years.
It's real interesting if you look at where we draw from, I don't think we have one single per son in clerical from Worthington or Dublin, too rich. Now we have a fair few from Westerville and the east side, they come around on the outer- belt. The rest come from Columbus or the rural areas away from the city.
Among the other most frequently given reasons for relocat ing to Worthington or Westerville were: cheaper land and adjacent land for future expansion; affordable space in an industrial park with the infrastructure already in place; new land that the firm could "make it's mark on" (by way of architectural design and sign-posting); and proximity to the owner's or president's home.
However it could be argued that an implicit acceptance of spatial entrapment underpins many of the techniques that 219 the ten firms utilized when extending their employee search to the external labor market. Although seven of the firms use Columbus newspapers to advertise openings, three of the personnel managers said that their firms have begun to advertise just in the local suburban newspapers and two firms also advertise in the papers of the surrounding rural areas, specifically to attract a particular type of female employee. Five of the firms said that they get many of their employees through referrals, whereby present employ ees encourage their friends, relatives and neighbors to apply for a position that they know is available.93 One personnel manager explained why this technique was espe cially attractive to his firm.
We do advertise in the paper, and that's how we get a lot of applicants, but we've found that the word-of-mouth referrals have a much higher stay ing rate. I think about thirty percent of our new employees are referred through word-of-mouth, but they represent seventy percent of the employ ees who stay with us for more than one year. The opposite is true where we have about sixty per cent come in to us from the newspaper ads, but they represent eighty percent of the people who leave within three months... it's because of the social nature of it, if their friend works here and they get positive reinforcement about it being a nice place to work, they stay longer. The newspaper people may be more wage oriented, and that's not a strong point here, also they may feel they don't fit in.
93 Indeed thirteen of the women interviewed, or almost half, found their jobs in this manner. 220
6.8 SUMMARY
This chapter has presented an interpretation of twenty- eight in-depth interviews conducted with female clerical workers living in either Westerville or Worthington. After an encapsulation of the spatial entrapment thesis, illus trations from the interviews were employed to highlight the spuriousness of linking women's short commutes with heavy domestic reponsibilities. It was shown that the situations of the women interviewed are more complicated than a simple immobilization created by their attempts to negotiate mul tiple roles. This chapter offered an alternative explana tion of the women's journeys-to-work to that of the spatial entrapment thesis. Drawing on the conceptualizations pre sented in Chapters 2 and 3, this alternative explanation partly consists of an examination of the changing powers and needs of women in paid employment in recent decades.
These changing powers and needs can either facilitate or impede women's attempts to negotiate their multiple roles.
It was recognized that the interviewees have the distinct ability to plan and learn new ways of acting and so possi bly transform both their powers and needs, and their socio- spatial system. Hence attention focussed on how the inter viewees have created coping strategies and even altered their powers and needs so as to better negotiate their mul tiple needs. Indeed it was noted that the women themselves 221 have been making attempts to better negotiate their various roles, by modifying their own expectations, by employing domestic help, and by altering their domestic division of labor.
While the women interviewed actively alter their socio- spatial system, their ability to do so is clearly mediated by wider changes in society. Thus changes in the infra structure of institutions and practices have enabled more women to better negotiate their multiple roles. It was noted that two such changes which have taken place in the recent past are the increase in child-care facilities and the apparently increased flexibility of employers towards their female employees. This was followed by a discussion of how to explain women's journeys-to-work. Attention focussed on how the women's commutes reflect their changing powers and needs, the spatial patterns of job opportuni ties, and the extent to which they are enmeshed into an existing set of localized relations.
The results of interviews with the personnel managers of ten firms were also presented. It was reported that prox imity to cheap female labor was not a particularly impor tant criterion in the relocation decision of most of the firms. Again, the situation is more complex than is implied by the literature. This also suggests that the suburbanization of office work to exploit spatially 222 entrapped women is, to say the least, somewhat of an over generalization. CHAPTER VII
SUMMARY AND CONCLUSIONS
7.1 SUMMARY OF THE MAJOR FINDINGS
This dissertation has used a variety of data to examine the suburbanization of female clerical workers in the
Columbus metropolitan area. Maps were employed to deter mine whether or not this had actually occurred. The first set of maps, employing Census shares of female clerical workers, provided some initial support for this. However, when the maps were re-estimated to account for the changing distribution of women of (paid) labor force age (by employ ing location quotients) the patterns seen in the shares maps all but disappeared. This shed some doubt on the ini tial conclusion that there had been a decentralization of female clerical workers in Franklin County.
Regression analyses of tract shares of female clerical workers on distance and socio-economic and housing vari ables were presented. These suggested that the changing pattern of female clerical workers over time was closely associated with the changing spatial distribution of women but that this relationship weakens somewhat after 1960.
- 223 - 224
Distance from the C.B.D. became increasingly important, changing it's sign from negative to positive, indicating
that more suburban locations were experiencing increases in
female clerical employment independently of the changing
geography of women of paid labor force age. It was also
found that shares of female clerical workers became
decreasingly related to childlessness. This tied in with
the literature on the increased paid employment of mothers
with very young children. Turning to the housing vari
ables, it was found that the effect of recently constructed
housing did become progressively more important, as Nelson
suggested. The same could not be said of home-ownership or
median family income, however. Neither performed as
expected.
The chapter also presented studies of the journeys-to-
work of clerical workers who live and/or work in a number
of Columbus' suburbs. The results for data for female
clerical workers drawn from the Polk's Suburban Directory
provided only very limited support for the spatial entrap
ment aspect of Nelson's argument. Examining data on the
zip-code of residence of clerical workers employed at three
companies located in Worthington or Westerville generally
strengthened doubts about the universality of the spatial
entrapment thesis. While large numbers of the clerical
workers did reside fairly close to their work-places, there 225 were also large numbers of women travelling from all over the city, and from distant areas outside of Franklin Coun ty. Finally in Chapter 5 the results of a survey question naire were presented. These largely confirm the over generalized nature of the spatial entrapment thesis.
Besides undertaking lengthy commutes, the categories of respondents conventionally viewed to be the most entrapped
(for instance, mothers with young children) actually had long journeys-to-work. On the other hand, women who, con ventionally, should have been less spatially entrapped
(non-mothers and women with adult children, for instance) were found to have short journeys-to-work. This suggests that, at least for the Columbus area, the evidence is not especially strong with regard to women being trapped in their place of residence, waiting for employment opportuni ties to come their way.
Chapter 6 presented an interpretation of twenty-eight in-depth interviews with female clerical workers living in either Westerville or Worthington. These interviews gener ally illustrate the spuriousness of linking women's short commutes with heavy domestic reponsibilities. The situ ations of the women interviewed are more complicated than a simple immobilization created by their attempts to negoti ate multiple roles. An alternative explanation of the women's journeys-to-work was offered. Drawing on the con 226 ceptualizations presented in Chapters 2 and 3, this alter native explanation partly consisted of an examination of the changing powers and,needs of women in the paid employ ment in recent decades. These changing powers and needs can either facilitate or impede women's attempts to negoti ate their multiple roles. It was recognised that the interviewees have the distinct ability to plan and learn new ways of acting and so possibly transform both their powers and needs, and their socio-spatial system. Hence the main thrust of the chapter was an investigation of how the interviewees have created coping strategies and even altered their powers and needs so as to better negotiate their multiple needs. For instance, the women spoke of modifying their own expectations, employing domestic help, and altering their domestic division of labor.
While the women interviewed actively alter their socio- spatial system, their ability to do so is clearly mediated by wider changes in society. Thus changes in the infra structure of institutions and practices have enabled more women to better negotiate their multiple roles. It was noted that two important changes which have taken place in the recent past are the increase in child-care facilities and the apparent increased flexibility of employers towards their female employees. These changes are considered in the next two sections. 227
The results of interviews with the personnel managers of ten firms were also presented. It was reported that prox imity to cheap female labor was not an important criterion in the relocation decision of most of the firms. Again, the situation is more complex than is implied by the liter ature. This also suggests that the suburbanization of office work to exploit spatially entrapped women is, to say the least, somewhat of an over-generalization.
The women interviewed dealt with their multiple roles in different ways, hence the emphasis in this dissertation on the individual women's coping strategies. Clearly there are different sub-groups among the women interviewed, each sub-group actually facing different situations from one another: divorced women with young children, women near retirement, single women, married women with teenage chil dren and so on. By definition these women have different powers and needs and susceptibilities, which will be exer cised or suffered in different ways: most divorced women with children economically have a need for paid work with full benefits for their family just to survive, whereas this need is not as pressing for married women, if the fam ily is covered under her partner (although it appears to have become increasingly necessary for families to have two incomes to maintain a certain stardard of living). Accord
ingly these different powers and needs imply different 228 strategies: part-time work with no benefits is not really a viable option for divorced women, especially if they have children, whereas it might be for married women.
This dissertation has attempted to capture some of the complexity and intricacies of the everyday lives of women who happen to be clerical workers. It has illustrated that there is no one-to-one relationship between women's multi ple roles and space: we cannot simply read off the spatial from what are basically stereotypical images of the role conflicts of women. Clearly women who are in paid employ ment are presented with many possibilities. They weigh up their various needs, along with their circumstances when deciding what coping strategies to employ. Thus spatial outcomes will always vary because people do not react mechanistically. Indeed people have the distinct ability to learn new ways of acting and planning, and so can change their circumstances.
7.2 CONTRIBUTIONS OF THIS DISSERTATION
This section places the research in this dissertation into broader context and evaluates it's contributions to urban and social geography. There appear to be four major contributions, two of an empirical character, and two which are more epistemological. First, this research has con tributed to a growing body of empirical research which highlights the mobility differences between women and men and among women. Second, this work adds substance to the literature situated at the intersection of the spatial divisions of labor, "industrial" location and female paid employment literatures. Although other studies have bridged these literatures, they have tended to be at the regional level and almost exclusively of British locations.
Third, this research has been an application of extensive and intensive research to a concrete research problem.
This is a contribution in itself as there have been few efforts that actually go beyond conceptualization in the attempts to forge a link between social theory and human geography. Furthermore, while there are numerous examples of extensive research (see for example Murgatoyd and Urry,
1985; Walby, 1985; 1986) there are no examples of realist inspired research which employed both extensive and inten sive research on the topic of women's employment. Fourth this dissertation contributes to the growing effort by feminist geographers to build a sound theory on which to base geographic research on gender relations.
During the initial phases of feminist geography the main efforts were to expose the sexist, male-as-normal bias in geography and render visible the previously invisible women
in geography. Thus the first feminist geography research was involved with correcting stereotypes, careless omis sion, and providing descriptive studies of the differences between women's and men's spatial behavior. More recently attention has begun to focus on the limitations of struc tural approaches to feminist geography which cast women into the role of helpless victims hindered by and trapped in space. This dissertation has exposed the tendency to assign space a spurious causal status. As an alternative, this dissertation, informed by social theory, forwarded the view that research on women should proceed in a manner that conceptualizes them as being active in both society and space with the aim of learning how they view the world, the decisions they make and how they cope with the wider social forces which are imposed on them. Appendix A
INTERVIEW ITEMS FOR INTERVIEWS WITH WOMEN
A. YOUR FAMILY:
1. What is your marital status?
Married before?
2. Do you have any children who live with you?
IF YES, how many, what are their sex and their ages?
3. Who else lives with you as part of your household?
B. YOUR JOB:
1. Who do you work for?
2. In general what is their line of business?
3. What is your job title?
4. How long have you worked there?
5. a.} How many hours a week do you work?
b.) How many hours a day do you work?
c.) What hours do you work?
- 231 - 232
6. Why did you choose this work schedule?
7. How did you get this job?
C. YOUR EMPLOYMENT HISTORY;
1. Did you have jobs before this one?
2. What type of job/hours/length of time in each job?
3. How did you get the jobs?
4. Why did you leave these jobs?
5. How was this related to your marriage and family?
(eg interrupted by child birth, did not intend to work
after marriage, compare with HUSBAND'S work history)
6. What was your first job when you went back to work
after having your children?
D. YOUR JOURNEY-TO-WORK:
1. How do you travel to work?
2. How many cars are there in your household?
3. I need to calculate the distance you travel to work, 233
can you give me the address of the firm you work for?
4. How long does it take you to get to/from work (TOTAL
time from door-to-door)?
5. How easy or difficult do you think it would be to
find a similar job in this area with better pay?
6. If you left this job for any reason what do you think
would be the possibility of finding another one?
7. How long have you lived at this address?
8. Have you lived anywhere else since working at your
present job?
IF YES, what were the names of your street and nearest
intersecting street?
9. Does you journey to work include dropping off your
child(ren) at a day-care center or school?
a.) IF YES, what is the address (or name of street
and the nearest intersecting street)?
b.) IF YES, did this affect your decision to work
where you presently work rather than somewhere else
further away?
10. How far are you prepared to travel to work? 234
E. YOUR EMPLOYER:
1. Why this particular employer?
2. Does the firm you work for have a lot of employees?
3. What is the ratio of men to women at work?
4. Do you like working for a big/small firm and why?
5. Why do you work in the private sector/for the state?
6. Why do you think you were hired?
7. What qualifications/personal attributes do you think
were considered?
8. What are employers looking for?
9. Do employers have biases? e.g. marital status, family
obligations, social background?
10. What is the general atmosphere like at work?
Do you get time off to care for your children when
they are sick, does your employer "look after you"?
11. Is there anything your employer could do to make it
easier for you to work, or that employers in general
could do for working women/mothers?
12. What do you particularly like/dislike about your job 235
and working there?
F. CLERICAL WORK:
1. Why do you go out to work?
2. Why clerical work?
3. Have you thought about working at home? Why/why not?
4. If you became unemployed and were looking for another
job what would you be looking for?
work which you have done before
good pay and benefits
hours which suit your needs
promotion prospects
pleasant people to work with
interesting/demanding work
something other than clerical work
a job where you could acquire new skills
G. YOUR FAMILY BACKGROUND:
1. What was your father’s occupation? 236
2. When you were growing up, did your mother ever go
out to work?
IF YES, what was her occupation
3. Has this had an effect on you? Made you want to go out
to work?
4. Has your family background influenced you in your
decision to work, or attitudes about women (and
mothers) working?
H. ATTITUDE TO JOB:
I. How do you feel about your job?: is it a career, a
temporary source of household income...
2. What does your salary contribute to the household
income? makes life more pleasant (ie. not vital)?
needed to pay mortgage? children's school fees?...
3. If you stopped working, what type of things would you
and your family have to spend less on or go without?
4. Is waged work a life-long commitment for you, or has
it become one? Why/why not?
5. Has your self-image/identity changed since working?
6. Why not other sources of identity — voluntary or charity work?
7. Did/has working change/changed your way of thinking,
your attitudes, your beliefs?:
e.g. do you find you have less in common now with
your non-working friends.
8. Has working led you to change your political views?
9. How do your husband/parents/children/friends feel
about you working?
e.g. some men don't want their wives to work.
10. Is it likely that you will change employers in
the future and why?
I. THE DUAL ROLE:
1. How do you manage the two aspects of your life (your
family and your job)?
household appliances (micro-wave, freezer)
child care (what type?)
domestic chores (paid help, family members)
chose job close to home
2. It has been suggested that household appliances make 238
it easier for women to work. Which of these do you own:
micro-wave
freezer
washing machine
vacuum cleaner
refrigerator
dish-washer
automatic oven
tumble-drier
Do you think they've made it easier for you to work?
Which of these did your mother have?
3. What sorts of things make it possible for you to work
(child-care, for instance)?
4. Why did you choose these options over others?
5. What types of adjustments have your family had to make
since you've been working (housework, cooking, laundry).
6. Describe how you allocate your time in a typical
working week 239
Do your husband and children help around the house?
IF YES, give examples of what they do.
7. Do you anticipate your present situation changing? i.e.
if you have children, when children leave home, if
you get married, when husband retires....
IF YES, in what ways and why?
Would/have you change(d) your own work/career plans to
fit in with your husband's?
J. CHILDREN AND CHILD CARE:
1. Child-care arrangements: why did you pick the
child-care arrangements that you have?
2. Hours of operation: did this influence you job choice?
3. How important are your relatives in helping relieve
the problems you face as a working woman?
4. What happens to the children during the school holidays?
5. Did you deliberately plan to start having children
when you did?
IF YES, did you ever think about how this would affect
your future working life? 240
6. Do your children contribute anything to the household
income?
7. Did you contribute anything.to your parent's household
income before you left home?
K. DUAL ROLE AND LOCATION:
1. How important is it that you work close to home (or
is it?) and why?
2. Would you like to work closer to/further from home?
3. Were there any jobs closer to/further from home that
you could have taken? Why didn't you take them?
4. Was work-place proximity important in residential choice?
5. Do you have any relatives in town who can help you out?
L. LABOR UNIONS:
1. Is there a labor union where you work?
2. Do you belong to the labor union (have you ever belonged
to one)?
IF YES, are/were you actively involved, attending
meetings, holding an offical position, been on strike? 241
IF NO, why did you decide not to join? Under what
circumstances would you join a union?
3. Does/did anyone in your family belong to a union?
4. How do you feel about labor unions?
5. How do people at work feel about labor unions?
6. What purpose do you think unions serve?
M. WOMEN'S ISSUES;
1. What do you think about women's issues?
child care facilities
equal pay
sex discrimination
sexual harassment on the job
maternity/parental leave
2. a.) Do you think legislation in these areas has/will
change your life?
b.) What about changing the lives of women in general?
c.) Do you think that such legislation was/is a good
or bad idea? Why do you think that? 242
3. What does your husband think about these issues?
4. Have you experienced increased disagreement about
women's issues with your husband/family/friends since
you began working again?
5. What are the attitudes of friends at work about
women's issues and politics?
6. If there has been a change in attitude on women's
issues has this been associated with, or led to,
changes in personal associations?
7. Would you like your daughter (if you had one) to have
the same experiences that you've had as a working
woman/mothe r ?
IF NOT, what do you hope will be different for her?
N. FINALLY:
1. What is the biggest problem that you face as a
working woman/mother?
2. What is the biggest problem for all working
women/mothe r s ?
3. Why do you think more women are going out to work
now than before? 243
OR
"Today more families find it necessary to have two
incomes (both the husband and wife's) in order to
have a decent standard of living."
Do you think this statement is true or false?
How do feel about this?
0. SOME INFORMATION ABOUT YOU:
1. What is your age?
2. How many years of education do you have?
3. Do you own or rent your home? Appendix B
INTERVIEW ITEMS FOR PERSONNEL MANGERS
What is the firm's general line of business?
How many people do you employ:
in total
in management
in clerical
How do these figures break down by gender, age, marital status, race, parental status?
What types of tasks do the clerical workers do?
What qualifications and qualities do you look for when considering hiring a clerical worker?
Have you had any problems with absenteeism or turnover?
If YES, what measures have you taken to limit this?
What do you think clerical employees find attractive and unattractive about working here?
What types of benefits do you offer clerical employees, 245
and do they vary by length of service, hours worked per
week and so on?
9. Is there a labor union?
If NO, what is the grievance procedure?
10. Do you face any special problems with hiring women?
Pregnancy policy, flex-time, child-care
Flex-time/part-time/job-share
Child-care facilities
Sick leave used to care for children
11. What types of recruiting strategies do you use?
Newspaper adverts
Internal labor markets
Word-of-mouth
12. What is the history of the firm's location here in
Westerville/Worthington?
13. Where do most of your clerical workers live?
How do they travel to work?
****Introduce the questionnaires, explain the need for
journey-to-work information. 246
****Ask if they have a zip-code listing of their clerical
workers' places of residence that they can give you. Appendix C
ADDITIONAL STATISTICS FROM THE REGRESSION
ANALYSES
Table 28
Means and Standard Deviations of Variables Employed
1950 1960 1970 1980 SHRFCL 1.031 0.680 0.483 0.398 (0.786} (0.451) (0.294) (0.217)
DIST 4.058 4.506 4. 757 5.161 (3.021) (2.835) (2.812) (2.993)
MEDINC 3491.700 5970.653 10042.388 29387.697 (1319.91) (2487.36) (3260.34) (26841.55)
SHRWOM 1.031 0.680 0.478 0.395 (0.516) (0.339) (0.257) (0.192)
SHRHSL10 1.124 0.704 0.478 0.395 (1.372) (0.828) (0.511) (0.503)
SHROWN 1.042 0.685 0.478 0.395 (0.629) (0.339) (0.299) (0.243)
SHRCHILD 1.031 0.680 0.478 0.395 (0.441) (0.404) (0.279) (0.335)
- 247 - Table 29
Zero Order Correlations
1950 1960 1970 1980 DIST -0.359 -0.152 -0.047 0.226
MEDINC -0.040 -0.106 -0.042 0.129
SHRWOM 0. 751 0. 773 0.824 0. 762
SHRHSL10 0.097 0.154 0.343 0.491
SHROWN 0. 536 0.569 0.531 0.569
SHRCHILD 0. 559 0.508 0.492 0. 560 Appendix D
SURVEY QUESTIONNAIRE
PLEASE FILL IN OR CIRCLE THE APPROPRIATE RESPONSE(S)
A. YOUR JOURNEY-TO-WQRK:
1. How do you travel to work?
walk/cycle/bus/drive with spouse/drive yourself/
car-pobl/motor bike/other(specify)
2. How long does it take you to get to/from work,
TOTAL time from door-to-door? ______minutes.
3. How many miles do you travel to/from work?
______miles.
4. Does your journey to work include dropping off your
child(ren) at a child-care center or school? YES/NO
IF YES, how many miles does this add to your to your
work trip? ______miles
How many minutes does this add to your work-trip?
______minutes
How long would your work trip be if you did not have
- 249 - 250
to drop your child(ren) off at the child-care
center/school ______(minutes and miles)
Did this affect your decision to work here rather than
somewhere further from home? YES/NO
5. Have you lived anywhere else since working here? YES/NO
IF YES, how many miles was it to work? ______miles
How long did it take you to get to work? ______minutes
B .YOU AND YOUR FAMILY;
1. What is your sex? male/female
2. What is your marital status?
single/married/divorced/separated/divorced
3. Have you been married more than once? YES/NO
If YES, how many times? 1 2 3 4 5
4. Do you have any children who live with you? YES/NO
If YES, how many, what is/are their sex and age(s)?
5. Do you have any children who have already left home?
YES/NO
If YES, how many, what is/are their sex and age(s)? 251
6. Who else lives with you as part of your household?
live alone/spouse/significant other/child(ren)/
parent(s)/parent(s )-in-law/lodger/other (specify)
7. Do you own or rent your home? OWN/RENT
8. How long have you lived at your present address?
______years
9. If married, does your spouse work? YES/NO
If YES, where and what is the address (to assess
her/his journey-to-work)
Name of Employer ______
Address ______
C. YOUR JOB:
1. What is your job title?
2. How many hours a week do you work?
3. What hours do you work? (eg. M-F 10-3; M W F 8-5)
4. How long have you worked here?
years ____ months 252
D. SOME INFORMATION ABOUT YOU:
1. What is your age?
21-25 26-30 31-35 36-40 41-45 46-50 51-55
56-60 61-65 65+
2. How many years of education did have you have in...
high school 1 2 3 4 5
college 0 12 3 4 5
3. When you were growing up, did your mother ever go out
to work? YES/NO
IF YES, what is/was her occupation?
4. What is/was your father's occupation? BIBLIOGRAPHY
Adler, P. 1983. "Rethinking Skill Requirements of New Tech nologies", Working Paper (9-784-027), Harvard Business School: Cambridge.
Aglietta, M. 1979. A THEORY OF CAPITALIST REGULATION, New Left Books: London.
Alexander, I., 1979. OFFICE LOCATION AND PUBLIC POLICY, Longman: New York.
Allen, E. L. and Edmonds, J. A. 1979. THE FUTURE OF THE PERSONAL AUTOMOBILE IN THE UNITED STATES, Institute for Energy Analysis: Oak Ridge, Tennessee.
Appelbaum, E. 1987. "Restructuring Work: Temporary, Part- Time, and At-Home Employment" in Hartmann, H. I. (ed.) COMPUTER CHIPS AND PAPER CLIPS: TECHNOLOGY AND WOMEN'S EMPLOYMENT, National Academy Press: Washington, D.C.
Armstrong, P. and Armstrong, H. 1983. A WORKING MINORITY: WHAT WOMEN MUST DO FOR PAY, Canadian Advisory Council for the Status of Women: Ottawa.
Armstrong, R. B. 1972. THE OFFICE INDUSTRY: PATTERNS OF GROWTH AND LOCATION, MIT: Cambridge.
Banner, M. G., Berheide, C. W. and Greckel, F. R. 1982. "Housing Preferences in Louisville: A Feminist Critique of the Suburban Environment", HOUSING AND SOCIETY, 9 (2): 95-110.
Baran, B. 1985. "Office Automation and Women's Work: The Technical Transformation of the Insurance Industry", in Castells M. (ed), HIGH TECHNOLOGY, SPACE, AND SOCIETY, Sage Publications: Beverly Hills.
Baran, B. 1987. "The Technical Transformation of White- Collar Work: A Case Study of the Insurance Industry" in Hartmann, H. I. (ed.) COMPUTER CHIPS AND PAPER CLIPS: TECHNOLOGY AND WOMEN'S EMPLOYMENT, National Academy Press: Washington, D.C.
- 253 - 254
Baron, R. D. and Norris, G. M. 1976. "Sexual Division and the Dual Labor Market", in Baker, D. L. and Allen, S., (eds.), DEPENDENCE AND EXPLOITATION IN WORK AND MARRIAGE, Longman: London.
Barrett, M. 1980. WOMEN'S OPPRESSION TODAY: PROBLEMS IN MARXIST FEMINIST ANALYSIS, New Left Books: London.
Barrett, N. S. 1979. "Women in the Job Market: Occupations, Earnings, and Career Opportunities", in Smith, R. E. (ed), THE SUBTLE REVOLUTION: WOMEN AT WORK, The Urban Institute: Washington, D. C.
Beechey, V. 1977. "Some Notes on Female Waged Labour in Capitalist Production", CAPITAL AND CLASS, 3:45-67.
Beechey, V. 1979. "On Patriarchy" FEMINIST REVIEW, 2:66-82
Beechey, V. 1985. "The Shape of the Work-force to Come" MARXISM TODAY.
Berk, R. and Berk, S. F. 1979. LABOR AND LEISURE AT HOME: CONTENT AND ORGANIZATION OF THE HOUSEHOLD DAY, Sage: Beverly Hills.
Berk, S. F. 1980. WOMEN AND HOUSEHOLD LABOR, Sage: Beverly Hills.
Berk, S. F. 1985. THE GENDER FACTORY: THE APPORTIONMENT OF WORK IN AMERICAN HOUSEHOLDS, Plenum Press: New York.
Berry, B. J. L., and Kasarda, J. D. 1977. CONTEMPORARY URBAN ECOLOGY, MacMiIlian: New York.
Beyer, J. 1985. "Geography of Women's Spaces", paper pre sented at the annual meetings of the Association of American Geographers, Detroit.
Bhaskar, R. 1975. A REALIST THEORY OF SCIENCE, Harvester Press: Brighton.
Bhaskar, R. 1979. THE POSSIBILITY OF NATURALISM, Harvester Press: Brighton.
Bianchi S. M. and Spain D. 1986. AMERICAN WOMEN IN TRANSITION, Russell Sage Foundation: New York.
Blau, F. D. 1984. "Discrimination Against Women: Theory and Evidence", in Darity, W. A. (ed.), LABOR ECONOMICS: MODERN VIEWS, Kluwer-Nijhoff: Boston. 255
Blau, F. D. and Ferber, M. A. 1986. THE ECONOMICS OF WOMEN, MEN, AND WORK, Prentice-Hall: Englewood Cliffs, New Jer sey.
Blau, F. D. and Jusenius, C. L. 1976. "Economists' Approaches to Sex Segrgation in the Labor Market: An Appraisal", SIGNS, 1 (3): 827-855.
Bluestone, B., and Harrison, B. 1986. "The Great American Jobs Machine: The Prolification of Low-Wage Employment in the U.S. Economy", Report prepared for the U.S.,Con gressional Joint Economic Committee.
Bluestone, B., Tilly, C. and Harrison, B. 1986. "Raising Inequality", in Obey, D. R. and Sarbanes, P., THE CHANGING AMERICAN ECONOMY, Blackwell: New York.
Bloom, D. E. 1982. "What's Happened to the Age at First Birth in the United States? A Study of Recent Cohorts", DEMOGRAPHY, 19: 351-370.
Bloom, D. E. and Trussell, J. 1984. "What are the Determi nants of Delayed Child-bearing and permanent Childless ness in the United States?" DEMOGRAPHY, 21: 591-611.
Bonefield, W. 1987. "Reformulation of State Theory", CAPITAL AND CLASS, 33: 96-127.
Bose, C., Feldberg, R., Sokoloff, N., (with the Women and Work Research Group) (eds.), 1988. HIDDEN ASPECTS OF WOMEN'S WORK, Praeger: New York.
Bose, C. 1982. "Technology and Changes in the Division of Labour in the American Home", in Robertson, M., (ed.), THE CHANGING EXPERIENCE OF WOMEN,
Bowlby, S., Foord, J. and McDowell, L. 1986. "The Place of Gender in Locality Studies", AREA, 18(4): 327-331.
Bowlby, S.R., Foord, J. McDowell, L. and Momsen, J. 1982. "Environment, Planning and Feminist Theory: A British Perspective", ENVIRONMENT AND PLANNING, A, 14(6): 711-716.
Bowlby, S. and McDowell, L. n.d. "The Feminist Challenge to Social Geography", Unpublished Mimeo, University of Reading.
Bowles, S. and Gintis, H. 1982. "The Crisis of Liberal Dem ocratic Capitalism: The Case of the United States" POLITICS AND SOCIETY, 11 (3): 51-93. 256
Boyer, R. 1979. "Wage Formulation in Historical Perspec tives The French Experience", CAMBRIDGE JOURNAL OF ECONOMICS, 3.
Boys, J. 1984. "Is there a Feminist Analysis of Architec ture?", BUILT ENVIRONMENT, 10:25-34.
Bradbury, K. L. 1986. "The Shrinking Middle-Class" NEW ENGLAND ECONOMIC REVIEW, (September/October).
Braverman, H. 1974. LABOR AND MONOPOLY CAPITALISM, Monthly Review Press: New York.
Brooker-Gross, S. R. and Maraffa T. A. 1985, "Commuting Distance and Gender among Non-Metropolitan University Employees", PROFESSIONAL GEOGRAPHER, 37 (3): 303-310.
Brownhill, S. 1984. "From Critique to Intervention: Socialist-Feminist Perspectives on Urbanization", ANTIPODE 16 (3): 21-34.
Bruegal, I. 1979. "Women as a Reserve Army of Labour" FEMINIST REVIEW, 3: 12-23.
Burris, V. 1982. "The Dialectic of Women's Oppression: Notes on the Relation between Capitalism and Patriarchy" BERKELEY JOURNAL OF SOCIOLOGY, 27: 51-74.
Cain, G. C. 1966. MARRIED WOMEN AND THE LABOR FORCE, Uni versity of Chicago: Chicago.
Cervero, R. 1986. SUBURBAN GRIDLOCK, Center for Policy Research: New Brunswick, N.J.
Cherlin, A. J. 1981. MARRIAGE, DIVORCE AND REMARRIAGE, Har vard University Press: Cambridge. -
Cichocki, M. K. 1980. "Women's Travel Patterns in a Subur ban Development", in Wekerle, G. R., Peterson, R. and Morley, D. (eds.), NEW SPACE FOR WOMEN, Westover Press: Boulder.
Clark, D. 1984. POST-INDUSTRIAL AMERICA: A GEOGRAPHICAL PERSPECTIVE, Methuen: New York.
Cockburn, C. 1984. BROTHERS: MALE DOMINANCE AND TECHNICAL CHANGE, Pluto Press: London.
Cook, C. and Rudd, N. 1984. "Factors Influencing the Resi dential Location of Female Householders", URBAN AFFAIRS QUARTERLY, 20: 78-96. 257
Corcoran, M. and Duncan, G. J., 1978. "Work History, Labor Force Attachment and Earning Differences Between Races and Sexes", THE JOURNAL OF HUMAN RESOURSES, 14 (1): 3-20.
Coser, R. L. 1982. "Stay Home, Little Sheba: On Placement, Displacement, and Social Change", in Kahn-Hut, R., Kaplan Daniels, A. and Colvard, R., WOMEN AND WORK: PROBLEMS AND PERSPECTIVES, University Press: New York.
Coulson, M., Magas, B., and Wainwright, H. 1975. "’The Housewife and her Labour under Capitalism1: A Critique", NEW LEFT REVIEW, 89: 59-71.
Cowen, R. S. 1974. "A Case Study of Technical and Social Change: The Washing Machine and Working Life", in Hart mann, H. and Banner, L. W. (eds.) CLIO'S CONSCIOUSNESS RAISED: NEW PERSPECTIVES ON THE HISTORY OF WOMEN, Harper and Row: New York.
Cox, K. R. and England, K. V. L. 1988. "Education and Resi dence in Historical Perspective", unpublished Mimeo, Department of Geography, The Ohio State University.
Coyle, A, 1982. "Sex and Skill in the Organisation of the Clothing Industry" in West, J., (ed.), WORK, WOMEN AND THE LABOUR MARKET, Routledge and Kegan Paul: London.
Craig, C. J., Rubery, J., Tarling, R., and Wilkinson, F. 1982. LABOUR MARKET STRUCTURE, INDUSTRIAL ORGANIZATION AND LOW PAY, Cambridge University Press: Cambridge.
Crompton, R. 1986. "Women and the 'Service Class'" in Crompton, R. and Mann, M., (eds.) GENDER AND STRATIFICATION, Polity Press: Oxford and New York.
Crompton, R. and Jones, G. 1984. WHITE COLLAR PROLETARIAT: DESKILLING AND GENDER IN CLERICAL WORK, London: Macmil- lian Press Ltd.
Crompton, R., Jones, G., and Reid, S. 1982. "Contemporary Clerical Work: A Case Study of Local Government", in West, J., (ed), WORK, WOMEN AND THE LABOUR MARKET, Rout- ledge and Kegan Paul: London
Crompton, R. and Reid, S. 1982. "The Deskilling of Clerical Work", in Wood, S. (ed.), THE DEGRADATION OF WORK: SKILL, DESKILLING AND THE LABOUR PROCESS, Hutchinson: London. 258
Cubukgil A. and Miller E. J. 1982. "Occupational Status and the Journey-to-Work", TRANSPORTATION, 11: 251-276.
Currie, E., Dunn, R . , and Fogarty, D. 1980. "The New Immis- eration: Stagflation, Inequality, and the Working Class", SOCIALIST REVIEW, 10 (6): 7-32.
Daniels, P. W. 1974. "New Offices in the Suburbs" in John son, J. H. (ed.) New York: John Wiley & Sons. SUBURBAN GROWTH, New York: John Wiley & Sons.
Daniels, P. W. 1979. SPATIAL PATTERNS OF OFFICE GROWTH AND LOCATION, Wiley: London.
Daniels, P. W. 1982. "An Exploratory Study of Office Loca tion Behavior in Greater Seattle" URBAN GEOGRAPHY, 3 (1): 58-78.
Daniels, P. W. 1985. SERVICE INDUSTRIES: A GEOGRAPHICAL APPRAISAL, Methuen: London.
Davidoff P. 1977. "Review of Barry Schwartz, THE CHANGING FACE OF THE SUBURBS", JOURNAL OF THE AMERICAN INSTITUTE OF PLANNERS, 43.
Davies, M. 1975. "Women's Place is at the Typewriter: the Feminization of the Clerical Labor Force", in Edwards, R., Reich, M. and Gordon, D., (eds.), LABOR MARKET SEGMENTATION, D. C. Heath: Lexington, MA.
Davies, M. W. 1982. WOMEN'S PLACE IS AT THE TYPEWRITER: OFFICE WORK AND OFFICE WORKERS, 1870-1930, Temple Uni versity Press: Philadelphia.
Davis, M. 1984. "The Political Economy of Late-Imperial America", NEW LEFT REVIEW, 143: 6-38.
Davy, T. 1986. "'A Cissy Job for Men: A Nice Job for Girls": Women Shorthand Typists in London, 1900-1939", in Davidoff, L. and Westover, B., (eds.), OUR WORK, OUR LIFES, OUR WORDS: WOMEN'S HISTORY AND WOMEN'S WORK, Mac- Millian: London.
Della Costa, M. 1975. "Women and the Subversion of the Com munity", in Della Costa, M. and James, S., (eds.) THE POWER OF WOMEN AND THE SUBVERSION OF THE COMMUNITY, Falling Wall Press: Bristol.
De Vroey, M. 1984. "A Regulation Approach Interpretation of the Contemporary Crisis", CAPITAL AND CLASS, 23: 45-66. 259
Dex, S. 1985. THE SEXUAL DIVISION OF LABOUR: CONCEPTUAL REVOLUTIONS IN THE SOCIAL SCIENCES, Harvester: Brighton.
Dick, B., and Morgan, G. 1987. "Family Networks and Employ ment in Textiles", WORK, EMPLOYMENT AND SOCIETY, 1 (2): 225-246.
Dix, M. C., Carpenter, S., Clarke, M. I., Pollard, J. and Spencer, M. 1983. CAR USE, A SOCIAL AND ECONOMIC STUDY, Gower: Farnborough, England.
Doeringer, P. and Piore, M. 1971. INTERNAL LABOR MARKETS AND MANPOWER ANALYSIS, D.C. Heath: Lexington, MA.
Ehrenreich, E. B. and English, D. 1975. j:The Manufacture of Housework", SOCIALIST REVOLUTION, 5 (4): 5-40.
Dunford, M. F. 1979. "Capital Accumulation and Regional Development in France", GEOFORUM, 10: 81-108.
Eisenstein, Z. 1979. CAPITALIST PATRIARCHY AND THE CASE FOR SOCIALIST FEMINISM, Monthly Review Press: New York.
England, K. V. L. 1985. "Conflict over Education in a Spa tial Context: A Critical Review", paper presented at the annual meetings of the Association of American Geogra phers, Detriot.
England, K., Jonas, A. and Mair, A. 1987. "The Practical Adequacy of METHOD IN SOCIAL SCIENCE", ANTIPODE, 19 (2): 240-254.
England, P. and Farkas, G. 1986. HOUSEHOLDS, EMPLOYMENT AND GENDER: A SOCIAL, ECONOMIC, AND DEMOGRAPHIC VIEW, Aldine Publishing Company: New York.
Ericksen, J. A. 1977. "An Analysis of the Journey to Work for Women", SOCIAL PROBLEMS, 24: 428-435.
Ettorre, E. M. 1978. "Women, Urban Social Movements and the Lesbian Ghetto" INTERNATIONAL JOURNAL OF URBAN AND REGIONAL RESEARCH, 2 (3).
Evans, M and Ungerson, C. 1983. SEXUAL DIVISIONS: PATTERNS AND PROCESSES, Tavistock Publications: London and New York.
Ewen, S., 1976. CAPTAINS OF CONSCIOUSNESS, Random House: New York. 260
Fagnani, J. 1983. "Women's Commuting Patterns in the Paris Region", TYDSCHRIFT VOOR ECONOMISCHE EN SOCIAALE GEOGRAFIE, 74: 12-24.
Fava, S. 1980. "Women's Place in New Suburbia", in Wekerle, G., Peterson, R . , Morley, D., (eds.), NEW SPACE FOR WOMEN, Westview: Boulder.
Feldberg, R. L. and Glenn, E. N. 1982. "Male and Female: Job Versus Gender Models in the Sociology of Work", in Kahn-Hut, R., Kaplan Daniels, A., and Colvard, R. (eds.), WOMEN AND WORK: PROBLEMS AND PERSPECTIVES, Oxford University Press: New York.
Feldberg, R. L. and Glenn, E. N. 1988. "Technology and Work Degradation: Effects of Office Automation on Women Clerical Workers", in Bose, C., Feldberg, R., Sokoloff, N., (with the Women and Work Research Group), (eds.), HIDDEN ASPECTS OF WOMEN'S WORK, Praeger: New York.
Finch, J. 1983. MARRIED TO THE JOB: WIVES INCORPORATION IN MEN'S WORK, Allen and Unwin: London.
Finch, J. 1984. "'It's Great to have Someone to Talk to': The Ethics and Politics of Interviewing Women", in Bell, C. and Roberts, H. (eds), SOCIAL RESEARCHING: POLITICS, PROBLEMS, PRACTICE, Routledge and Kegan Paul: Boston.
Fincher, R. 1987. "Social Theory and The Future of Urban Geography", PROFESSIONAL GEOGRAPHER, 39 (1): 9-12.
Foord, J. and Gregson, N. 1986. "Patriarchy: Towards a Reconceptualization" ANTIPODE, 18 (2): 186-211.
Fox, M. , 1983. "Working Women and Travel: The Access of Women to Work and Community Facilities", JOURNAL OF THE AMERICAN PLANNING ASSOCIATION, 49 (2): 156-170.
Fox, M. and Hesse-Biber, S. 1984. WOMEN AT WORK, Mayfield: Palo Alto, California.
Freeman, J. 1981. "Women and Urban Policy" in WOMEN AND THE AMERICAN in Stimpson, C. R., Dixler, E., Nelson, M., and Yatrakis, K. B., (ed), WOMEN AND THE AMERICAN CITY, Uni versity of Chicago Press: Chicago.
Freeman, R. 1986. "Factor Prices, Employment and Inequality in a Decetralized Labor Market", National Bureau of Eco nomic Research/Corporation for Enterprise Development, (June). 261
Fuchs, V. R. 1971. "Differences in Hourly Earnings Between Men and Women", MONTHLY LABOR REVIEW, 91 (5): 9-15.
Gardiner, J. 1975. "Women's Domestic Labour", NEW LEFT REVIEW, 89s 47-58.
Gerson, K. 1983. "Changing Family Structure and the Posi tion of Women", JOURNAL OF THE AMERICAN PLANNING ASSOCIATION, 49 (2)s 138-148.
Giddens A. 1979. CENTRAL PROBLEMS IN SOCIAL THEORY: ACTION, STRUCTURE, AND CONTRADICTION IN SOCIAL ANALYSIS, MacMil- lian: London.
Giddens, A. 1981. A CONTEMPORARY CRITIQUE OF HISTORICAL MATERIALISM, VOL 1: POWER, PROPERTY AND THE STATE,
Giddens A. 1985. "Time, Space and Regionalisation", in Gregory, D. and Urry, J. (eds.), SOCIAL RELATIONS AND SPATIAL STRUCTURES, St. Martin's Press: New York.
Gittus, D. 1986. "Marital Status, Work and Kinship, 1850-1930." in Lewis, J., (ed.), LABOUR AND LOVE: WOMEN'S EXPERIENCE OF HOME AND FAMILY, Basil Blackwell: Oxford.
Glazer, N. 1980. "Everyone Needs Three Pairs of Hands: Doing Unpaid and Paid Work", in Berk, S. F., (ed.), WOMEN AND HOUSEHOLD LABOR, Sage: Beverly Hills.
Glenn, E. N. and Feldberg, R. L. 1982. "Degraded and Des killed: The Proletarianization of Clerical Work", in Kahn-Hut, R., Kaplan Daniels, A. and Colvard, R., (ed), WOMEN AND WORK: PROBLEMS AND PROSPECTS, Oxford Universi ty Press: New York
Glenn, E. N. and Feldberg, R. L. 1984. "Clerical Work: The Female Occupation", in Freeman, J. (ed), WOMEN: A FEMINIST PERSPECTIVE, Mayfield: Mountainview, CA.
Goldberg, R. 1983. ORGANIZING OFFICE WORKERS: DISSATISFACTION, CONSCIOUSNESS AND ACTION, Praeger: New York.
Gordon, I and Molho, I. 1985. "Women in the Labour Markets of the London Region: A Model of Dependence and Con straint" URBAN STUDIES, 22: 367-386.
Gordon, L. 1977. WOMAN'S BODY, WOMAN'S RIGHT: A SOCIAL HISTORY OF BIRTH CONTROL IN AMERICA, Viking: New York. 262
Graff, C. L. 1982. "Employment of Women, Suburbanization and Housing Styles", HOUSING AND SOCIETY, 9 (2): 111-117.
Gregory, D. 1985. "Suspended Animation: The Stasis of Dif fusion Theory", in Gregory, D. and Urry, J, (eds.), SOCIAL RELATIONS AND SPATIAL STRUCTURES, St. Martin's Press: New York.
Gruntfest, E. 1988. "Geographic Perspectives on Women: A Status Report", paper presented at the annual meetings of the Association of American Geographers, Phoenix.
Guiliano, G. 1979. "Public Transportation and the Travel Needs of Women" TRAFFIC QUARTERLY, 33: 607-616.
Gurin, D. B. 1981. "Women Future Travel, Driving and Tran sit Use", PSYCHOLOGYOF WOMEN QUARTERLY, 6: 72-83.
Hagerstrand, T. 1967. INNOVATION DIFFUSION AS A SPATIAL PROCESS, University of Chicago Press: Chicago.
Hagerstrand, T. 1970. What about the People in Regional Science?, PROCEEDINGS OF THE REGIONAL SCIENCE ASSOCIATION, 24.
Hall, P. 1985. "The Geography of the Fifth Kondratieff", in Hall, P. and Markusen, A. SILICON LANDSCAPES, Allen and Unwin: Boston.
Hanson, S and Hanson, P. 1980. "Gender and Urban Activity Patterns in Uppsala, Sweden", GEOGRAPHICAL REVIEW, 70: 291-299.
Hanson, S. and Hanson, P. 1981. "The Travel-Activity Pat terns of Urban Residents: Dimensions and Relationships to Sociodemographic Characteristics", ECONOMIC GEOGRAPHY, 57: 332-347.
Hanson, S and Johnston, I. 1985. "Gender Differences in Work-Trip Length: Explanations and Implications" URBAN GEOGRAPHY, 6 (3): 193-219.
Harris, L. 1983. "Industrialization, Women and Working Class Politics an the West of Ireland", CLASS AND CAPITAL No. 19.
Harrison, B., Tilly, C., and Bluestone, B. 1986. "Rising Inequality", in Obey, D, R. and Sarbanes, P., THE CHANGING AMERICAN ECONOMY, Blackwell: New York. 263
Hartgen D. T. 1978. "Can Current Transportation Planning Methods Analyze Women's Travel Issues?" in WOMEN"S TRAVEL ISSUES: RESEARCH NEEDS AND PRIORITIES, U.S. Department of Transportation, Conference and Proceed ings: Washington, D.C.
Hartman-Strom, S. 1987. "'Machines Instead of Clerks': Technology and the Feminization of Bookkeeping, 1910-1950, in Hartmann, H. I. (ed.) COMPUTER CHIPS AND PAPER CLIPS: TECHNOLOGY AND WOMEN'S EMPLOYMENT, National Academy Press: Washington, D.C.
Hartmann, H. 1976. "Capitalism, Patriarchy, and Job Segre gation by Sex", SIGNS, 3: 137-168.
Hartmann, H. 1979. "The Unhappy Marriage of Marxism and Feminism: Towards a More Progressive Union", CAPITAL AND CLASS, 8: 1-33.
Hayden, D. 1981. THE GRAND DOMESTIC REVOLUTION, M.I.T. Press: Cambridge.
Hayford, 1974. "The Geography of Women: An Historical Introduction", ANTIPODE, 6: 1-19.
Hillier, B. and Hanson, J. 1984. THE SOCIAL LOGIC OF SPACE, Cambridge University Press: Cambridge.
Hofferth, S. L. 1984. "Some Long Term Economic Consequences for Women of Delayed Childbearing and Reduced Family Size." DEMOGRAPHY, 21: 141-155.
Hogan, D. 1978. "Education and the Making of the Chicago Working Class, 1880-1930" JOURNAL OF THE HISTORY OF EDUCATION, : 227-269.
Holcomb, B. 1984a, "Women in the City", URBAN GEOGRAPHY, 5 (3), 247-253.
Holcomb, B. 1984b. "Women in the Rebuilt Urban Environment: The United States Experience", BUILT ENVIRONMENT, 10 (1), 18-24.
Holcomb, B. 1986. "Geography and Urban Women", URBAN GEOGRAPHY, 7 (5): 448-456.
Howe, A. and O'Conner, K. 1982. "Travel to Work and Labor Force Participation of Men and Women in an Australian Metropolitan Area", PROFESSIONAL GEOGRAPHER, 34. 264
Hudson, R. 1980. "Regional Development Policies and Female Employment"AREA, 12 (3) 229-234.
Humphries, J. 1977. "Class Struggle and the Persistence of the Working Class Family", CAMBRIDGE JOURNAL OF ECONOMICS, Is 241-258.
Hunt, H. A. and Hunt, T. L. 1987. "Recent Trends in Cleri cal Employments The Impact of Technilogical Change", in Hartmann, H. I. (ed.) COMPUTER CHIPS AND PAPER CLIPS: TECHNOLOGY AND WOMEN'S EMPLOYMENT, National Academy Press: Washington, D.C.
Hunt, P. 1980. GENDER AND CLASS CONCIOUSNESS, MacMillian: London.
Issel, W. 1979. "Americanization, Acculturation and Social Control: -School Reform Ideology in Industrial Pennsyl vania, 1880-1910", JOURNAL OF SOCIAL HISTORY, Fall: 569-590.
Jones, E. F. 1981. "The Impact of Women's Employment on Marital Fertility in the US, 1970-75", POPULATION STUDIES, 35: 161-173.
Kaniss, P. and Robins, B. 1974. "The Transportation Needs of Women." in Hapgood, K. and Getzels, J., (eds), PLANNING, WOMEN AND CHANGE, American Society of Planning Officials: Chicago.
Katz, M.B. 1977. CLASS, BUREAUCRACY AND SCHOOLING: THE ILLUSION OF EDUCATIONAL CHANGE IN AMERICA, Vintage Books: New York.
Keller S. 1981. BUILDING FOR WOMEN, Lexington Books: Lex ington, Massachusetts.
Kennedy, S. E. 1979. IF ALL WE DID WAS TO WEEP AT HOME: A HISTORY OF WHITE WORKING-CLASS WOMEN IN AMERICA, Indiana University Press: Bloomington.
Kessler-Harris, 1981. WOMEN HAVE ALWAYS WORKED: A HISTORICAL OVERVIEW, The Feminist Press: Old Westbury.
Kessler-Harris, 1982. OUT TO WORK: A HISTORY OF WAGE-EARNING WOMEN IN THE UNITED STATES, Oxford Univer sity Press: New York.
Klein, E. 1984. GENDER POLITICS, Harvard University Press: Cambridge. 265
Kocka, J. 1980. WHITE COLLAR WORKERS IN AMERICA, 1890-1940: A SOCIO-POLITICAL HISTORY IN INTERNATIONAL PERSPECTIVE, Sage: Beverly Hills.
Kosters, M. H., and Ross, M. N., 1987. "The Influence of Employment Shifts and New Job Opportunities on the Growth and Distribution of Real Wages", in CONTEMPORARY ECONOMIC PROBLEMS, American Enterprise Institute.
Lamphere, L. 1985. "Bringing the Family to Work: Women's Culture on the Shop Floor", FEMINIST STUDIES, 11 (3): 519-540.
Larwood, L. and Gutek, B. A. 1985. "Women at Work in the USA", in Davidson, M. J. and Copper, C. L. (eds.) WORKING WOMEN: AN INTERNATIONAL SURVEY, John Wiley: New York.
Lasch, C. 1977. HAVEN IN A HEARTLESS WORLD, Basic Books: New York.
Lawrence, R. Z. 1984. "Sectoral Shifts and the Size of the Middle-Class", THE BROOKINGS REVIEW, 3 (1): 3-11.
Levy, F. 1987. DOLLARS AND DREAMS: THE CHANGING AMERICAN INCOME DISTRIBUTION, Russell Sage Foundation: New York.
Ley, D. 1977. "Social Geography and the Taken-for-Granted World", TRANSACTIONS OF THE INSTITUTE OF BRITISH GEOGRAPHERS, 2: 498-512.
Lewis, J. 1984. "The Role of Female Employment in the Industrial Restructuring and Regional Development of the Post-War U.K.", ANTIPODE 6 (3): 47-59.
Lewis, J. 1985. "The Debate on Sex and Class", NEW LEFT REVIEW, 149: 108-120.
Lewis, J. and Foord, J. 1984. "New Towns and Gender Rela tions in Old Industrial Regions: Women's Employment in Peterlee and East Kilbride", BUILT ENVIRONMENT, 10 (1): 42-52.
Lipietz A. 1986. "New Tendencies in the International Divi sion of Labor: Regimes of Accumulation and Modes of Reg ulation", in Scott, A. J. and Storper, M. , (eds.), PRODUCTION, WORK, TERRITORY: THE GEOGRAPHICAL ANATOMY OF INDUSTRIAL CAPITALISM, Allen and Unwin: Winchester.
Littler, C. 1982. "Deskilling and Changing Structures of Control", in Wood, S. (ed.), THE DEGRADATION OF WORK: 266
SKILL, DESKILLING AND THE LABOUR PROCESS, Hutchinson: London.
Lockwood, D. 1958. THE BLACKCOATED WORKER, Allen and Unwin: London.
Loveman, G. W . , and Tilly, C. 1988. "Good Jobs or Bad Jobs: What Does The Evidence Say?", NEW ENGLAND REVIEW, (January/February): 46-65.
Loyd, B. S., Monk, J. J. and Rengert, A. C. 1982. "Land scapes of the Home" in Monk, J. J. and Rengert, A. C. (eds.), WOMEN AND SPATIAL CHANGE: LEARNING RESOURSES FOR SOCIAL SCIENCE COURSES, Kendall and Hunt: Dubuque, Iowa.
MacKenzie, S. 1980. "Women's Place - Women's Space: A Per spective on the Geographical Study of Women", AREA, 12: 47-49.
Mackenzie, S. 1984. "Catching Up with Ourselves: Ideas on the Developing a Gender-Sensitive Theory in the Environ mental Disciplines, 6(3): 16-18.
MacKenzie, S. 1984. "Editorial Introduction", ANTIPODE, 16 (3): 3-10.
MacKenzie, S. 1987. "Women's Responses to Economic Restruc turing: Changing Gender, Changing Space" in Hamilton, R. and Barrett, M. (ed), THE POLITICS OF DIVERSITY: FEMINISM, MARXISM AND NATIONALISM, Verso: London.
MacKenzie, S. and Rose, D. 1983. "Industrial Change, the Domestic Economy and Home Life", in Anderson, J. Dun can, S., Hudson, R., (eds.), REDUNDANT SPACES? SOCIAL CHANGE AND INDUSTRIAL DECLINE IN CITIES AND REGIONS, Academic Press: London.
McDowell, L. 1983. "Towards an Understanding of the Gender Division of Urban Space" ENVIRONMENT AND PLANNING D, 1 (1): 59-72.
McDowell, L. and Massey, D. 1984. "A Women's Place" in Mas sey, D. and Allen, J. (eds.) GEOGRAPHY MATTERS: A READER, Cambridge University Press: Cambridge.
McNally, F. 1979. WOMEN FOR HIRE: A STUDY OF THE FEMALE OFFICE WORKER, MacMiIlian: London.
McDonough, R. and Harrison, R. 1978. "Patriarchy and Rela tions of Production", in Kuhn, A. and Wolpe, A. (eds.), FEMINISM AND MATERIALISM: WOMEN AND MODES OF PRODUCTION, Routledge and Kegan Paul: London. 267
Mackintosh, M. 1977. "Reproduction and Patriarchy", CAPITAL AND CLASS, 2: 119-127.
Mackintosh, M. 1979. "Domestic Labour and the Household", in Burman, S. (ed), FIT WORK FOR WOMEN, Croom Helm: Lon don.
Mackintosh, M. 1981. "The Sexual Division of Labour and the Subordination of Women", in Young, K., Wolkowitz, C. and McCullagh, R., OF MARRIAGE AND THE MARKET: WOMEN'S SUBORDINATION IN INTERNATIONAL PERSPECTIVE.
Mallier, A. T. and Rosser, M. J. 1987. WOMEN AND THE ECONOMY: A COMPARATIVE STUDY OF BRITAIN AND THE USA, MacmiIlian: London.
Madden, J. F. 1981. "Why Women Work Close to Home", URBAN STUDIES, 18: 181-194.
Madden, J. F. and White, M. J. 1980. "Spatial Implications of Increases in the Female Labor Force: A Theoretical and Empirical Synthesis", LAND ECONOMICS, 56 (4).
Malos, E. 1980. (ed.) THE POLITICS OF HOUSEWORK, Alison and Busby: London.
Manners, G 1974. "The Office in the Metropolis: an opportu nity for shaping metropolitan America" ECONOMIC GEOGRAPHY, 50.
Markusen, A. R. 1981. "City Spatial Structure, Women's Household Work, and National Urban Policy" in Stimpson, C. R., Dixler, E., Nelson, M., and Yatrakis, K. B., (eds.), WOMEN AND THE AMERICAN CITY, University of Chi cago Press: Chicago.
Marx, K. 1977. CAPITAL VOLUME ONE: A CRITIQUE OF POLITICAL ECONOMY, Vintage: New York.
Masnick, G. and Bane, M. J. 1980. THE NATION'S FAMILIES: 1960-1990, Joint Center for Urban Studies of M.I.T. and Harvard University: Cambridge.
Mason K. O., Czajka, J. L. and Arber, S. 1976. "Changes in the U.S. Women's Sex-Role Attitudes, 1964-1974", AMERICAN SOCIOLOGICAL REVIEW 4: 573-96.
Massey, D. 1983. "Industrial Restructuring, as Class Restructuring: Production Decentralisation and Local Uniqueness", REGIONAL STUDIES, 17 :73-89. 268
Massey, D. 1984. SPATIAL DIVISIONS OF LABOR: SOCIAL STRUCTURES AND THE GEOGRAPHY OF PRODUCTION, Methuen: New York.
Massey, D. 1985. "New Directions in Space", in Gregory, D. and Urry, J., (ed), SOCIAL RELATIONS AND SPATIAL STRUCTURES, St. Martin's Press: New York.
Massey, D. 1987. "The Shape of Things to Come", in Peet, R., (ed.), INTERNATIONAL CAPITALISM AND INDUSTRIAL RESTRUCTURING: A CRITICAL ANALYSIS, Allen and Unwin: Boston.
Matrix, 1984. MAKING SPACE: WOMEN AND THE MAN-MADE ENVIRONMENT, Pluto: London.
Mattaei, J. A. 1982. AN ECONOMIC HISTORY OF WOMEN IN AMERICA: WOMEN'S WORK, THE SEXUAL DIVISION OF LABOR, AND THE DEVELOPMENT OF CAPITALISM, Schocken Books: New York.
Matteis, R. 1979. "The New Back Office Focusses on Customer Service", HARVARD BUSINESS REVIEW, March-April.
Mazey, M.E. and Lee, D. 1983. "Her Space, Her Place: A Geography of Women", Association of American Geogra phers: Washington D.C.
Meiksins-Wood, E. 1988. "Capitalism and Human Emancipa tion", NEW LEFT REVIEW, 167: 1-21.
Milkman, R. 1976. "Women's Work and the Economic Crisis: Some Lessons of the Great Depression", REVIEW OF RADICAL POLITICIAL ECONOMICS, 8 (2): 73-97.
Miller, R. 1983. "The Hoover in the Garden: Middle-Class Women and Suburbanization, 1850-1920", ENVIRONMENT AND PLANNING D, 1: 73-87.
Molyneux, M. 1979. "Beyond the Domestic Labour Debate", NEW LEFT REVIEW, 116: 3-27.
Murgatroyd, L. and Urry, J. 1985. "The Class and Gender Restructuring of the Lancaster Economy, 1950-1980" in Murgatroyd, L., Savage, M., Shapiro, D., Urry, J., Wal- by, S., Warde, A., with Mark-Lawson, J. (eds.) LOCALITIES, CLASS AND GENDER, Pion: London.
Monk, J. J. and Rengert, A. C. 1982. WOMEN AND SPATIAL CHANGE: LEARNING RESOURSES FOR SOCIAL SCIENCE COURSES, Kendall and Hunt: Dubuque, Iowa. 269
Murphree, M. C. 1987. "Not-So-Changing World of the Secre tary", in Hartmann, H. I. (ed.) COMPUTER CHIP'S AND PAPER CLIPS: TECHNOLOGY AND WOMEN’S EMPLOYMENT, National Academy Press: Washington, D.C.
9-to-5, National Association of Working Women, 1985. HIDDEN VICTIMS: CLERICAL WORKERS, AUTOMATION, AND THE CHANGING ECONOMY, 9-to-5, National Association of Working Women: Cleveland.
9-to-5, National Association of Working Women, 1986. WORKING AT THE MARGINS: PART-TIME AND TEMPORARY WORKERS IN THE UNITED STATES, 9-to-5, National Association of Working Women: Cleveland.
Nelson, K. 1986a. "Labor Demand, Labor Supply and the Sub urbanization of Low-Wage Office Work", in Scott, A. and Storper, M. (eds), PRODUCTION, WORK AND TERRITORY, Allen and Unwin: Winchester, MA.
Nelson, K. 1986b. "Automation, Skill and Back Office Loca tion", paper presented at the annual meeting of the Association of American Geographers, Minneapolis.
Netter, E. and Price, R. G. 1983. "Zoning and the Nouveau Poor", JOURNAL OF THE AMERICAN PLANNING ASSOCIATION, 49 (2), 171-181.
Nussbaum, K. 1982. OFFICE WORK IN AMERICA, Working Women, a National Association of Office Workers: Cleveland.
Oakley, A. 1974. THE SOCIOLOGY OF HOUSEWORK, Martin Robert son : London.
Oakley, A. 1981. "Interviewing Women: A Contradiction in Terms", in Roberts, H., (ed.), DOING FEMINIST RESEARCH, Routledge and Keagan Paul: London.
Ohio Senate Task Force on Women Single Heads of Households, 1985. SUMMARY REPORT, State House (Neal F. Zimmers): Columbus.
Oppenheimer, V. K. 1970. THE FEMALE LABOR FORCE IN THE UNITED STATES: DEMOGRAPHIC AND ECONOMIC FACTORS GOVERNING ITS GROWTH AND CHANGING COMPOSITION, Universi ty of California: Berkeley.
Palm, R. and Pred, A. 1978. "The Status of American Women: A Time-Geographic View" in Lanegran, D. A. and Palm, R., (ed.), AN INVITATION TO GEOGRAPHY, 2nd edition, McGraw- Hill: New York. 270
Pahl, R. 1984. DIVISIONS OF LABOUR, Blackwell: Oxford.
Pahl, R. and Wallace, C. 1985. "Household Work Strategies in Economic Recession", in Redclift, N. and Mingione, E., (eds.), BEYOND EMPLOYMENT: HOUSEHOLD, GENDER AND SUBSISTENCE, Basil Blackwell: Oxford.
Parcel, T. L. and Mueller, C. W. 1983. ASCRIPTION AND LABOR MARKETS, Academic Press: New York.
Pearce, D. 1983. "The Feminization of Ghetto Poverty", SOCIETY (November/December): 70-74.
Perrons, D. 1986. "Unequal Integration in Global Fordism: The Case of Ireland", in Scott, A. J. and Storper, M. (eds.), PRODUCTION, WORK, TERRITORY: THE GEOGRAPHICAL ANATOMY OF INDUSTRIAL CAPITALISM, Allen and Unwin: Win chester .
Petit, P. 1986. SLOW GROWTH AND THE SERVICE ECONOMY, Fran cis Pinter: London.
Phillips, A. and Taylor, B. 1980. "Sex and Skill: Notes Towards A Feminist Economics", FEMINIST REVIEW, 6:79-88
Pickup, L. 1984. "Women's Gender-Role and its Influence on their Travel Behaviour", BUILT ENVIRONMENT, 10 (1), 61-68.
Piore, M. 1986. "Beyond Social Anarchy", in Obey, D. R. and Sarbanes, P., THE CHANGING AMERICAN ECONOMY, Blackwell: New York.
Power, M, 1983. "From Home Production to Wage Labor: Women as a Reserve Army of Labour", REVIEW OF RADICAL POLITICAL ECONOMICS, 15 (1): 71-91.
Power, M. 1988. "Women, the State, and the Family in the US: Reaganomics and the Experience of Women", in Rubery, J., (ed.), WOMEN AND RECESSION, Routledge and Kegan Paul: New York.
Pred, A. 1985. "The Social Becomes the Spatial, the Spatial Becomes the Social: Enclosures, Social Change and the Becoming of Places in the Swedish Province of Skane" in Gregory, D. and Urry, J. (eds.), SOCIAL STRUCTURES AND SPATIAL STRUCTURES, St. Martin's Press: New York.
Preston, S. H. 1984. "Children and the Elderly: Divergent Paths for America's Dependents", DEMOGRAPHY, 16: 1-25. 271
Pucher, J., Hendrickson, C., and McNeil, S., 1981. "The Socioeconomic Characteristics of Transit Riders: Some Recent Evidence", TRAFFIC QUARTERLY, 45: 461-483.
Rapoport, R. and Rapoport, R. 1976. DUAL CAREER FAMILIES RE-EXAMINED, Martin Robertson: London.
Ravetz, A. 1984. "The Home of Women: A View of the Interi or", BUILT ENVIRONMENT, 10 (1): 8-17.
Robinson, J. P. 1980. "Housework Technology and Household Work" in Berk S. F., (ed.), WOMEN AND HOUSEHOLD LABOR, Sage: Beverly Hills.
Roessner, J. D., Mason, R. M., Porter, A. L., Rossini, F. A., Scwartz, A. P., and Nelms, K. R. 1985. THE IMPACT OF OFFICE AUTOMATION ON CLERICAL EMPLOYMENT, 1985-2000: FORCASTING TECHNIQUES AND PLAUSIBLE FUTURES IN BANKING AND INSURANCE, Quorum Books: Westpoint, Conn.
Roistacher, E. A. and Young, J. S., 1980. "Working Women and City Structure: Implications of the Subtle Revolu tion" SIGNS, 5 (3): S220-S225.
Roos, P. A. and Reskin, B. F. 1984. "Institutional Factors Contributing to Sex Segregation", in Reskin, B. F., (ed.), SEX SEGREGATION IN THE WORKPLACE: TRENADS, EXPLANATIONS, REMEDIES, National Academy Press: Washing ton, D.C.
Rose, D. 1984. "Re-thinking Gentrification: Beyond the Uneven Development of Marxist Urban Theory", ENVIRONMENT AND PLANNING D, 1 :47-74.
Rose, D. and Villeneuve, P. 1985. "Women and the Changing Spatial Division of Labour in Montreal, 1971-1981", paper presented at the Annual Meeting of the Association of American Geographers, Detroit.
Rosenthal, N. H. 1985. "The Shrinking Middle Class: Myth or Reality", MONTHLY LABOR REVIEW, 108 (March): 3-10.
Rotella, E. J. 1981. FROM HOME TO OFFICE: U.S. WOMEN AT WORK, 1870-1930, University of Michigan Research Press: Ann Arbor.
Rothblatt, D. N . , Garr, D. J. and Sprague, J. 1979. THE SUBURBAN ENVIRONMENT AND WOMEN, Praeger: New York.
Rubery, J. and Wilkinson, F. 1979. "Notes on the Nature of the Labor Process in the Secondary Sector", in LOW PAY 272
AND LABOUR MARKET SEGMENTATION (Conference Papers, Cam bridge) .
Rutherford B. M. and Wekerle, G. R. 1988. "A Model of An Urban Captive Labor Market: Causes and Consequences in Metropolitan Toronto", paper presented at the annual meetings of the Association of American Geographers, Phoenix.
Ryan, M. P. 1979. WOMANHOOD IN AMERICA, 2nd Edition, New Viewpoints: New York.
Rytina, N. F. 1982. "Tenure as a Factor in the Male-Female Earnings Gap", MONTHLY LABOR REVIEW, 105: 32-36.
Rytina, N. F. and Bianchi, S. M. 1984. "Occupational Reclassification and Changes in Distribution by Gender", MONTHLY LABOR REVIEW, 107: 11-17.
Saegart, S. 1981. "Masculine Cities and Feminine Suburbs: Polarised Ideas, Contradictory Realities" in Stimpson, C. R., Dixler, E., Nelson, M. 0., and Yatrakis, K, B. (eds.) WOMEN IN THE AMERICAN CITY, University of Chica go: Chicago.
Sanderson K. 1986. "'A Pension to Look Forward to ...?': Women Civil Service Clerks in London, 1925-1939", in Davidoff, L. and Westover, B. OUR WORK, OUR LIFES, OUR WORDS: WOMEN'S HISTORY AND WOMEN'S WORK, MacMiIlian: London.
Sapiro, V. 1986. WOMEN IN AMERICAN SOCIETY: AN INTRODUCTION TO WOMEN'S STUDIES, Mayfield: Palo Alto, California.
Sassen-Koob, S. 1984. "The New Labor Demand in Global Cit ies", in Smith, M. P. (ed.) CITIES IN TRANSFORMATION: CLASS, CAPITAL AND THE STATE, Sage: Beverly Hills.
Sayer, A. 1984. METHOD IN SOCIAL SCIENCE: A REALIST APPROACH, Hutchinson Ltd.: London.
Sayer, A. 1985. "The Difference that Space Makes" in Grego ry, D. and Urry, J. (eds.), SOCIAL STRUCTURES AND SPATIAL STRUCTURES, St. Martin's Press: New York.
Sayer, A. and Morgan, K. 1985. "A Modern Industry in a Declining Region: Links Between Method, Theory and Pol icy" in Massey, D. and Meagan, R., (eds) POLITICS AND METHOD, Methuen: London. 273
Schultz, T. W. 1974. "Marriage, Family Human Capital, and Fertility", supplement to THE JOURNAL OF POLITICLA ECONOMY, 82.
Seccombe, W. 1974. "The Housewife and her Labour Under Cap italism" NEW LEFT REVIEW, 83: 3-24.
Selwood, D. 1982. "Office Location within a Metropolitan Area", unpublished Mimeo, Department of Geography, The Ohio State University.
Selwood, D. 1985. "Labor Processes and the Location of Offices", unpublished Mimeo, Department of Geography, Ohio State University.
Sharpe, S. 1984. DOUBLE IDENTITY: THE LIVES OF WORKING WOMEN, Penguin: Harmondsworth, England.
Shapiro, B. 1982. "Women Facing Special Housing Problems", HOUSING AND SOCIETY, 9 (2), 67-69.
Shepard, J. M. 1971. AUTOMATION AND ALIENATION, M.I.T. Press, Cambridge, MA.
Sirbu, M. A. 1982. "Understanding the Social and Economic of Office Automation", Unpublished Mimeo, M.I.T.: Cam bridge .
Smith R. E. (ed.) 1979. THE SUBTLE REVOLUTION: WOMEN AT WORK The Urban Institution: Washington, D.C.
Smith, J. P. and Ward, M. P. 1985. "Time-Series Growth in the Female Labor Force", JOURNAL OF LABOR ECONOMICS, 3.
Smith, V. 1983. "The Circular Trap: Women and Part-Time Work", BERKELEY JOURNAL OF SOCIOLOGY, 18: 1-17.
Soja, E. W. 1985. "The Spatiality of Social Life: Towards a Transformative Retheorisation" in D. Gregory and J. Urry (ed.), SOCIAL STRUCTURES AND SPATIAL STRUCTURES, St. Martin's Press: New York.
Sokoloff, N. J. 1987. "What's Happening to Women's Employ ment: Issues for Women's Labor Struggles in the 1980's-1990's", in Bose, C., Feldberg, R., Sokoloff, N., (with the Women and Work Research Group), (eds.), HIDDEN ASPECTS OF WOMEN'S WORK, Praeger: New York.
Finch, J. 1983. MARRIED TO THE JOB: WIVES INCORPORATION IN MEN'S WORK, Allen and Unwin: London. Stanback, T., and Noyelle, T. 1982. CITIES IN TRANSITION, Allenheld: Totowa, N.J. 274
Sternlieb G, Hughes J. W. and Hughes C. 0. 1982. DEMOGRAPHIC TRENDS AND ECONOMIC REALITY: PLANNING AND MARKETS IN THE 80's, Center for Urban Policy Research: New Brunswick.
Strasser, S.M. 1980. "An Enlarged Human Existance? Technol ogy and Household Work in Nineteenth-Century America" in Berk, S. F., (ed.), WOMEN AND HOUSEHOLD LABOR, Sage: Beverly Hills.
Stolzenberg, R. M. and Waite, L. J. 1984. "Local Labor Mar kets, Children and Labor Force Participation of Wives", DEMOGRAPHY, 21 (2): 157-170.
Taylor, M. and Thrift, N. 1983. "Business Organization, Segmentation and Location" REGIONAL STUDIES, 17 (6): 445-465.
Thrift, N. and Pred, A. 1981. "Time-geography: A New Begin ning", PROGRESS IN HUMAN GEOGRAPHY, 5: 277-286.
Thrift, N. 1985. "Flies and Germs: A Geogrpaphy of Knowl edge", in Gregory, D. and Urry, J. (eds.), SOCIAL STRUCTURES AND SPATIAL STRUCTURES, St. Martin's Press: New York.
Thurow, L. C. 1987. "A Surge in Inequality", SCIENTIFIC AMERICAN, 256 (5): 30-37.
Tilly, L. A. and Scott, J. W. 1978. WOMEN, WORK, AND FAMILY Holt, Rineholt and Winston: New York.
Tivers, J. 1978. "How the Other Half Lives: The Geographi cal Study Of Women" AREA, 10: 301-396.
Tivers, J. 1985. WOMEN ATTACHED: THE DAILY LIVES OF WOMEN WITH YOUNG CHILDREN, Croom Helm: London.
Toft Jensen, H., Hansen, P. A., and Serin, G. 1983. "Capi talist Technology and the Change of the Labour Process", in Gillespie, A. TECHNOLOGICAL CHANGE AND REGIONAL DEVELOPMENT, Pion: London.
Treiman, D. J. and Terrell, K. 1975. "Women, Work, and Wag es: Trends in the Female Occupational Structure Since 1940", in Land, K. and Spilerman, S. (ed.) SOCIAL INDICATORE MODELS, Russell Sage: New York.
U.S. Department of Commerce, 1979. THE JOURNEY TO WORK IN THE U.S., 1975: CURRENT POPULATION REPORTS, Special Studies, P-23, No 99, U.S. Department of Commerce: Wash ington, D.C. 275
U.S. Department of Labor, 1980. PERSPECTIVES ON WORKING WOMEN: A DATA BOOK, Bureau of Labor Statistics: Wash ington, D.C.
Valli, L. 1986. BECOMING CLERICAL WORKERS, Routledge and Kegan Pauls Boston.
Van Horn, S. H. 1988. WOMEN, WORK AND FERTILITY, 1900-1986, New York University Press: New York.
Vanek, J. 1974. ’’Ti m e Spent on Housework", SCIENTIFIC AMERICAN, 31: 116-120.
Vanek, J. 1980. "Household Work, Wage Work and Sexual Equality", in Berk, S. F. (ed.) WOMEN AND HOUSEHOLD LABOUR, Sage: Beverly Hills.
Vogel, L. 1983. MARXISM AND THE OPPRESSION OF WOMEN: TOWARDS A UNITARY THEORY, Pluto: London.
Wainwright, H. 1984. "Women and the Division of Labour", in Abrams, P. and Brown, R., (eds), UK SOCIETY: WORK, URBANISM AND INEQUALITY, Weidenfeld and Nicolson: Lon don .
Waite, L. J. and Stolzenberg, 1976. "Intended Childbearing and Labor Force Participation of Young Women: Insights from Nonrecursive Models", AMERICAN SOCIOLOGICAL REVIEW, 41: 235-252.
Walby, S. 1985. "Spatial and Historical Variations in Wpmen's Unemployment and Employment", in Murgatroyd, L., Savage, M., Shapiro, D., Urry, J., Walby, S., Warde, A., with Mark-Lawson, J. (ed.) LOCALITIES, CLASS AND GENDER, Pion: London.
Walby, S. 1986. PATRIARCHY AT WORK: PATRIARCHAL AND CAPITALIST RELATIONS IN EMPLOYMENT, Polity Press: Cam bridge .
Wagner, P.K. 1984. "Suburban Landscapes for Nuclear Fami lies: The Case of Greenbelt Towns in the United States", BUILT ENVIRONMENT, 10 (1): 35-41.
Welter, B. 1966. "The Cult of True Womanhood: 1820-1860", AMERICAN QUARTERLY, Summer: 151-174.
Wekerle, G. 1981. "Women in the Urban Environment", in Stimpson, C. R., Dixler, E., Nelson, M., and Yatrakis, K. B., (eds), WOMEN AND THE AMERICAN CITY, University of Chicago Press: Chicago. 276
Wekerle, G. 1984. "A Woman's Place is in the City"/ ANTIPODE, 6 (3): 11-19.
Wekerle, G. R., Peterson R . , and Morley, D. 1980. NEW SPACE FOR WOMEN, Westview Press: Boulder.
West, J. 1982. WORK, WOMEN AND THE LABOUR MARKET, Routledge and Kegan Paul: London.
Westwood, S. 1984. ALL DAY EVERY DAY, Pluto: London.
Westoff, C. F. and Ryder, N. B. 1977. THE CONTRACEPTIVE REVOLUTION, Princeton University Press: Princeton.
Wheeler, J. A. 1967. "Occupational Status and Work-Trips: A Minimum Distance Approach", SOCIAL FORCES, 45 (4): 508-515.
Whickler, M. L. and Kronenfeld, J. J. 1986. SEX ROLE CHANGES TECHNOLOGY, POLITICS, AND POLICY, Praeger: New York.
White, M. J. 1977. "A Model of Residential Location Choice and Commuting By Men and Women Workers", JOURNAL OF REGIONAL SCIENCE, 17: 41-52.
Williamson, L. 1982. INDUSTRIAL RESTRUCTURING, LOCAL CLASS STRUCTURE AND FEMALE WAGED LABOUR ON MERSEYSIDE, Mimeo, Department of Urban and Regional Studies, University of Sussex.
Winckler, V. 1985. "Tertiarization and Feminization at the Periphery: The Case of Wales", in Newby, H., Bujra, J., Littlewood, P., Rees, G . , and Rees, T. L., (eds.), RESTRUCTURING CAPITAL: RECESSION AND REORGANIZATION IN INDUSTRIAL SOCIETY, Macmillan: London.
Winter, M. and Morris, E.W. 1982. "Housing Conditions, Sat isfaction and Conventionality: An Analysis of the Hous ing of Female Headed Households", HOUSING AND SOCIETY, 9 (2), 70-86.
Women and Geography Study Group of the IBG, 1984. GEOGRAPHY AND GENDER, Hutchinson: London.
Yeates, M. and Garner, B. 1980. THE NORTH AMERICAN CITY, Harper and Row: San Francisco.
Young, I. 1981. "Beyond the Unhappy Marriage: A Critique of the Dual Systems Theory", in Sargent, L. WOMEN AND REVOLUTION: A DISCUSSION OF THE UNHAPPY MARRIAGE OF MARXISM AND FEMINISM, South End Press: Boston. 277
Zelinsky, W. , Monk, J. and Hanson, S. 1982. "Women and Geography: A Review and Prospectus" in PROGRESS IN HUMAN GEOGRAPHY, 6 (3): 317-354,
Zelinzer, V. A. 1985. PRICING THE PRICELESS CHILD: THE CHANGING SOCIAL VALUE OF CHILDREN, Basic Books: New York.
Zorn, J. 1985. "Spatial Variations in the Gender Income Disparities", MODELING AND SIMULATION, 16 (Part 1): 145-148.